I have attended the Machine Learning seminar by Vinesh Kannan, who is working in Google but has had an internship in Civic Digital Fellowship this summer. His talk was about how they used Machine Learning in finding certain quantity of “job names” for companies in each industry.
He first talked about how job names are categorized, and the early stages of their research. As long as the quantity was similar to the market demand, the program was successful. But then they noticed that some of the required jobs were “washed away.” The reason why these happened was that these jobs were containing words that exist in various industries, like data analyst and economic analyst. He said that they resolved the problem by commanding that each job that contain such words will be accepted as different jobs, so that they won’t disappear.
As a freshman year, these were a little hard for me to keep up while listening and actually understand what he meant. Most of the things I wrote were from what I understood from my discussion with some of the audience after the seminar.
Machine Learning seems to be an interesting topic. Until now, I saw it applied as a program that picks words and predicts based on them; both in this seminar and previously in the DemonHacks as the winner back then used ML to predict health problems based on text messages.
IIT has a ML club (ML@IIT), and I want to be active in that community.
I attended Vinesh Kannan’s talk on Machine learning at the Bureau of labor statistics. Going in to the lecture I was curious on what type of work he did as I didn’t have any clue of what the BLS would be working on in terms of machine learning. They were trying to be able to aid the process in assigning different jobs soc codes that represent specific categories, and to do this they implemented machine learning. I think he said that there were like 10 million records that they had in their data set, which is a mindblowing number to me.
Although some of the talk was out of my scope, it was interesting to see how difficult it is to train these models and the weird problems that arise. Example of one was the term analyst appearing in many different fields so it essentially was being ignored in the models predictions, which led to wrong classifications. I learned a-lot about this type of stuff, and hopefully it will come in handy during my own personal machine learning projects. The fellowship seems like a very cool opportunity that I would love to do down the line, since machine learning is fascinating to me and I like the focus the work has on using technology for good. Overall I was really glad I have joined the ml@iit club as this has been one of many talks I have attended that were really fun and a good use of time in between classes! I will definitely be going to more as well
I attended a seminar on biochemical engineering by a professor at northwestern university, I didn’t know what the professor was going to talk about before I went to this seminar, because I’m a computer science major. And I’m afraid I don’t understand it at all. But when I arrived, my worries disappeared. The professor is a very good speaker, he can give a vivid explanation of knowledge, so that everyone can understand and think about it.
He first introduced his team, which also included his students. It was an excellent team. They explored a lot and had a lot of academic achievements. And then he talks a lot about academics.
For example, catalysis in energy processes, food can change state through the environment(like crispy taco and soft taco), how to establish a well-understood base case, microbial electrolysis, etc. I’m interested in all of them.
The seminar made me want to attend all kinds of seminars. Because I am curious about different things, it can enrich my knowledge, so I will actively participate in the seminar of various subjects in the future.
I attended a seminar on november 20th presented by Professor Justin Notestein, from the Dept. of Chemical and Biomolecular Engineering at Northwestern University, about active sites and how to manipulate them. I felt kind of scared at first because it was my first event here and I didn’t really know how things were going to go down so I was not sure if I was really going or not, but last minute I made up my mind and decided to go. And afterall it turned out to be a good experience in the end. Through the event I learned that you can put a metal on top of a catalyst to either cover the active sites completely or just cover the sites in which the bondage is less selective (as they will bond to the metal), what enables scientists to manipulate those catalysts to only synthesise certain things. This activity helped me understand how computer science can be used in scenarios in which I never thought they could. For example in this case computer science can be used to help them study the active sites of the catalysts and their properties in a more detailed way. Even though I’m not a big chemistry guy, I enjoyed the experience therefore I probably would attend more activities like that. The only thing I would advise is to have some background on the subject as it can sometimes be hard to understand part of the vocabulary that is being used.
On November 21st, I attended the Machine Learning seminar by Vinesh Kannan about his experiences working as a data science fellow at the Bureau of Labor Statistics. I was nervous before going to the seminar because I was a afraid of not being able to comprehend what’s the main point the talker is trying to express. But after arriving, I think that most of my anxiety lessen and I was able to concentrate on the talk. His main point in the seminar is that how him and his team used Machine Learning in finding job names for requested companies from different fields.
He goes in deeper using Machine on how different job names are separated into different categories suitable for the different purposes of research. Something new I learned from the talk is that for his team to success on produces occupation and wage data used by policymakers, hiring staff, job seekers, and researchers across the country, the quantity of job names has to be the same to the market demand. He goes on to mention that some of the required jobs seems to be unavailable and the reason is that a lot of industries also have key words associated with these jobs, they solved it by separating jobs that have those kinds of words be different jobs.
Machine learning was very helpful and interesting. Though I cannot grasp 100% of the entire seminar, I came back with a good sense of how Machine Learning works. I would attend these activities again.
I attended the machine learning seminar on November 21st by Vinesh Kannan. His seminar focused on his experience withing as a fellow at the Bureau of Labor Statistics specializing in data science. I was eager to go to the seminar because data science and machine learning are two things I’m very interested in and hope to study while completing my computer science degree. Vinesh’s talk was very professional and informative. He went into great detail about the type of work he did and how he completed it using advanced machine learning models and some of the largest labor-related data sets on the planet. I was especially intrigued by his explanation about how the Bureau of Labor Statistics categorized jobs at a very intricate level. Before attending this seminar, I wouldn’t have considered that such a massive amount of data needs to be collected on every working person and every job, and that it must be categorized in such a complex way. I also wouldn’t have guessed that machine learning as complex as Vinesh described would be implemented in order to do all of this work.
After attending this seminar, I’m still very interested in machine learning and data science, and applying for the data science fellowship with the Bureau of Labor Statistics is definitely something I plan on doing in the future.
I attended a seminar by Vinesh Kannan, who talked about Machine Learning. Before going to the seminar, I didn’t have much knowledge about Machine Learning. So this experience for me to learn something from Vinesh Kannan was an exciting experience. I will understand why most people will have been nervous when going into the seminar without any knownledge about Machine Learning. But me on the other hand I was nervous at first, which later turned into excitement because of how Vinesh Kannan described the field.
Vinesh Kannan explains what him and his team used Machine Learning to do. Kannan describes how his teams objective was to find certain jobs to a certain company. For the teams objective to be successful Kannan and his team had to separate the different jobs and those jobs had certain words that you can group them with similar jobs that have those same words. After the seminar was over my understanding of Machine Learning was much better than before the Seminar.
This Seminar was a great Seminar and I learned alot from Vinesh Kannan. I plan on going to more Seminars that are about different topics in the Computer Science field. I hope to use what ever I learned from the Seminars in the future.
I attended a seminar on november 21st presented by Plamen Petrov, Vice President, Artificial Intelligence and Exponential Technologies and Chief Data Officer at Anthem Inc., about Introducing AI into the digital-first healthcare delivery. When I was going to this event I felt interested and pretty excited about going to it, as the topic was really interesting and I was looking forward to learning more about that sometime this year. So I decided that it was a good time to start getting more involved with the topic. I learned that Thirty-five percent of U.S. adults have gone online to self-diagnose a medical condition that they or someone they knew had, and that multiple times they find things that are not accurate to what’s going on or find “false-answers” to their problems. So through this talk I learned that AI can be introduced to this, in a way in which they would learn the types of problems people are searching, the solution they found and if it worked or not, this way the AI would learn what problems people are facing and accurately deliver a solution to them. I would definitely attend another activity like this, as it was a really good experience and it got me more connected to the school and to computer science, especially if it is a topic that sparks interest in me or it is something that relates to something I want to learn or know about.
I attended the Machine Learning seminar hosted by Vinesh Kannan, an IIT Computer Science graduate on the 21st of November, who currently works at Google but shares his experiences working as a data science fellow at the Bureau of Labor Statistics (BLS). Before going to the seminar, I was nervous because I don’t have much experience in the CS-related field, and I won’t be able to comprehend and stay on the same page as the speaker. Still, at the same time, I was excited to learn more about his experience in data science and Machine Learning. As the seminar started, I became more and more comfortable with understanding the speaker’s thoughts and what he was trying to convey. His primary message in the seminar was about how he and his fellow team members used Machine Learning in finding job names and positions for different companies in different fields. He went deeper into his experience of using Machines to successfully sort the various categories and subcategories for different research purposes. I learned that Vinesh and his team used data on occupation and wages made by researchers, future employees, and employers. He told me about how data was used to measure the number of jobs and also how Machines were used to help the employers get their best matches for candidates for their positions. Even though I didn’t understand every aspect of what Vinesh was trying to convey, I learned about how Machine Learning can be used in the industrial side of things, and I would love to attend future events to learn more and expand my knowledge.
On November 21st, I attended a seminar presented by Mr. Plamen Petrov. Mr. Petrov is the Vice President, Artificial Intelligence and Exponential Technologies and Chief Data Officer of Anthem Inc. His seminar was about being a company that integrates Artificial Intelligence into healthcare. Before going to this event, I felt extremely excited about the concept as AI is a subject that I am super passionate about. Just the thought of integrating computing and technology into everyday life is mind blowing to me. Also, I was eager to learn more about the concept as I knew there was plenty for me left to learn. I learned that 35% of people in the Unites States research their symptoms online only to find the wrong diagnosis and a treatment that is not accurate for their true condition. I learned that through AI, people can connect their symptoms to those who have already been treated and get the proper diagnosis and treatment for their needs. I found that this way would be a lot more effective and seamless rather than having to deal with the wrong diagnosis more than 1/3 of the time. I am very happy to have gone to seminar like this and I would definitely attend another activity like this in the future. I felt as if this activity brought me closer to knowing more about my major and feel as if it is a catalyst for me to learn and know more things about the concept.
Before participating in the machine learning seminar activity, I didn’t really feel much other than getting ready to learn new things regarding data science, machine learning and much more. I felt quite confident that I would walk out with a series of loaded information.
Throughout this event I learned a lot about the internship known as the 2020 Civil Digital Fellowship program including the projects they attempted including data science, product management, and the software they used. I also learned that these students have made groundbreaking progress in terms of where they currently are. They pursued top graduate schools including many very prominent companies for work.
I will definitely attend an activity like this again. During my first semester, I haven’t really attended many talks from research professors or even speakers from different universities. However, after listening to this one, I have discovered that these events are very well performed and will provide a ton of information regarding the major you have selected and will pursue. This knowledge that one learns will be important for future use, so I might as well continue to attend talks like this to broaden my knowledge in this certain topic, and continue to better myself in my selected field.
I attended the seminar on 11/21/2019 by Plamen Petrov. This talk largely focused on the usage of AI in healthcare systems as a preventative tool. Before this talk, I honestly had very little background understanding of healthcare applications as I had never used one in the past, although I did have an interest in the AI aspect. One thing that I learned from the talk is how much well protected any information is pertaining to someones health information. For example, Petrov’s application was forced to use simulated data in some cases because of the strictness of these rules. My learning of computer science has been expanded as I had a previous interest in AI, so learning some of the struggles of producing an AI has further sparked my interests in AI.
I would also definitely like to attend another talk by Petrov. I personally think that it is important for anyone going into any field to hear and learn from someone who has been successful in it. To me, Petrov is the shining example of someone who has found success in computer science and AI systems, and has made strides to further enrich others planing on joining the field through his companies multiple college outreach programs and research opportunities.
I have also went to the Seminar by Plamen Petrov on Thursday, 11/12.29. I rarely go to seminars, but after attending one, my thoughts about it has changed. Before, I view seminars as long boring lectures with lots of information to memorize. There are lots of great ideas that need to be shared. This seminar focuses on making a algorithm about health. Specifically, one that could diagnose a patient without a doctor. This is a work in progress, but I could see that potential it can grow into. It would be an app one can download. Then the patients list symptoms, problems, or take pictures. This is actually harder than it looks and brings a whole new set of problems in ways I never have thought of. First, the algorithm needs to be able to analyze text and pictures. Because of the hundreds of ways a person can describe a symptom along with the massive data that requires to be effective, it becomes a challenge.
I was better than I had expected. I was expecting to fall asleep, but I learned a lot about the problems of having a algorithm to interact with people. I enjoyed this seminar and would go to a similar one.
Before going to the machine learning seminar I did not really attend any other lectures. I was very skeptical and thought it would be very boring. I was also very excited to learn more about cs and cs oriented stuff because I became a CS major with no prior experience with it. Throughout the seminar, I learned a lot about how the speaker made connections with other companies and how they landed internships and future jobs. The main point that I learned from the seminar was the ways that Kannan and his team members had to collect data and use it to help sort different job types with different job positions. One thing that stood out to me was when he mentioned using names to associate them with different job titles. When the seminar ended I was still a little confused about some of the things he talked about. I thought this was a good experience I will hopefully be attending more talks in the future.
Before attending Vinesh Kannan’s Seminar, I did not really know how I felt about the seminar. I did not completely understand what a fellowship is and I did not really know what he was doing at the fellowship. Listening to his experience with the fellowship really peaked my interest however as he had the opportunity to attend a government job, which covered his flights, living, and most expenses he had to take upon himself and he also got paid for the work he did. Although it was only a couple of weeks, he put his full-time job on hold to experience this fellowship. Him pushing back his job for this experience really shows how much importance this fellowship had to him and as he stated during the seminar, he did not regret doing that. He learned many things and got to explore a part of the country he has never been in. It was also interesting to learn about the fellowship itself, as he got to work at the Bureau of Labor Statistics, which oversees all the data we look at when we search up salaries based on different criteria. He explained to us how this job works and how they have slowly introduced machine learning that outputs the 6-digit code they have for each different job that is in the market. This was always done by hand and introducing machine learning made everything a lot more efficient and quicker. Although there were still plenty of mistakes, the human input in sorting the data grew a lot smaller and the data could be processed much quicker. This activity helped me understand CS a lot more as it showed me a real-life job that deals with coding on a daily basis and it also introduced me into fellowships that I may want to explore later on in my college career. I would happily attend another event like this as it gives me insight into the CS world and how it all works.
I went to the Machine Learning seminar on November 21st. The seminar was presented by Vinesh Kannan, a graduate from IIT. Before attending the seminar, I was unfamiliar and curious about machine learning. I wanted to better understand why machine learning is an important part of computer science. I am a computer science major, but I still do not know what I want to specialize in. Vinesh Kannan talked about his worked experience for the Bureau of Labor Statistics. He used machine learning to categorize different types of job titles for companies. He also talked about how it could be problematic if two jobs have identical words in their names, so he talked about how his team was able to separate similar results into separate job titles. After the seminar ended, I felt like I have learned a lot more about machine learning. This talk also expanded my interests in computer science and could possibly lead me into a certain career in the future. I was undecided about what career I wanted to pursue in the computer science field, but machine learning sounds like it is an important job to work in. I think that I would attend another computer science related seminar again.
On November 21st I attended a Machine Learning seminar talk by Vinesh Kannan. Beforehand, I was eager to go because machine learning is one of my favorite topics in cs. His seminar focused on his experience within as a fellow at the Bureau of Labor Statistics specializing in data science. The talk was very informative and professional. I already know a good amount of behind how machine learning works but I don’t have much experience with seeing how it is implemented to a practical application. Seeing the problem presented and then an explanation of what was used for training data, how certain problems were solved was very interesting. For example, solving the prediction unrealistic jobs by creating context models to correlate job title and industry. His talk definitely improved my understanding of machine learning and its practical uses in the industry. It was also very interesting to learn how the data he produces is used by policy makers to make decisions.
After attending this seminar my interest in machine learning continues. Although, since this specific field is not what I am interested in, I don’t think I will pursue the data science fellowship with the Bureau of Labor Statistics
On November 21st, I attended a talk on the current condition of AI’s reach into the healthcare field, hosted by Vice President Plamen Petrov of Anthem Inc. I was mildly apprehensive, even if a but curious, expecting another future-looking series of aspirations on how AI could improve the field “one day” but what I got was a reasonable, non-theoretical application of the technology.
Because of the current state of the Amercian Healthcare System and many Americans inability to afford decent coverage, Anthem Inc. seeks to build a simpler, more affordable alternative through the device that over 80% of Americans current possess: a smartphone. Their app, CareSpree, mimics a typical doctor-patient consultation and compares the subjective analysis of the patient to “People Like Me”: their categorization database comparing people of similar symptoms, backgrounds and ethnicity or race.
If nothing else, it taught me about the reaches of AI pattern recognition and its applications outside of facial recognition technology. Despite Computer Science being my major of choice, I can’t say I get much information about the current state of the AI field outside of wildly exaggerated articles. It’s already being put to test in imaging and video analysis to work towards dermatology and cardiology studies.
On Thursday, October 17, I attended the talk “Differential Privacy, Adaptive Data Analysis, and Free Speedups via Sampling” by mathematics associate professor Lev Reyzin from the University of Illinois at Chicago. When I saw the emails about the talk, I read the abstract and saw that the talk promised to show a “recently-discovered beautiful connection between differential private mechanisms and preventing overfitting in machine learning.” This immediately made me interested in attending it since I had some prior experience in machine learning from both a high school research project with computer vision (convolutional neural nets) and a high school CS class. Additionally, I had learned before about preventing overfitting in neural network-based models using a dropout layer and in decision tree-based models using pruning, so I was intrigued to find out what other methods there were to prevent overfitting in other machine learning models.
When I walked into the room for the talk, I felt instantly out of place. Most of the people in the room looked much older than me, and I assumed they were mostly graduate or doctoral students with specializations in AI. Nevertheless, I ignored the impostor syndrome and took a seat, opening up my notebook to jot down anything I could understand or find useful.
The talk was pretty difficult to understand mainly due to the mathematical notation he was using, which involved a lot of probability theory. He talked about switching from a static-based machine learning model to an adaptive-based one, and he went into depth about different types of queries (statistical, counting, etc.) that could be made to a model. I couldn’t quite grasp all of this, and to be honest, if you asked me about any of this, I probably wouldn’t be able to answer you with 100% confidence. However, what I took away from his talk (and what I found most interesting, coincidentally), was that you could prevent overfitting by applying a slight Laplace distribution error to the training dataset and subsampling it.
As someone who likes to dabble in a variety of CS-related fields, I’d be lying if I said I didn’t want to go to other CS talks after this slightly difficult experience. However, what I do know now is that I should probably choose one that is more suited towards my CS experience level! Perhaps a more intermediate talk on AI/ML would have been more productive for me, rather than this graduate-level one.
On November 21st, I attended a Machine Learning seminar talk by Vinesh Kannan at IIt. I was excited to go to this seminar because machine learning a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention. This intrigues me because I want to be able to learn more about machine learning so that one day I will be able to do it too. Vinesh Kannan was a graduate of IIT, which made me even more interested in his talk since I will also be a graduate from IIT one day. He focused his talk on how he and his colleagues use machine learning in finding different jobs and positions for different companies in all kinds of fields. He emphasized a lot how a lot of jobs are always done by hand and using machine learning it made everything a lot easier. He inspired me into learning about machine learning, he was also a great public speaker that inspired me not only on machine learning but also in helping communities around me. I will be attending a lot more seminars to learn more about how I can use computer science to make the world a better place.
I attended Vinesh Kannan’s seminar on Machine Learning. “Machine learning is the scientific study of algorithms and statistical models that computer systems use to perform a specific task without using explicit instructions” – Wikipidea. Most of what I knew about Machine Learning came from what I googled ten minutes before the seminar. To be honest, I was expecting it to just be another boring, mandated event we had to attend for this class. But that was not the case; after attending the event, I learned some new and intriguing information, and am even a little bit more fascinated by Machine Learning now. I was particularly interested in Kannan’s discussion about his experience at the Bureau of Labor Statistics, and how they would use a six digit code to generate information about different jobs, and how they logged all information about job salaries in their database as well. Another interesting topic Kannan brought up is how Machine learning will automate jobs that most people thought could only be done by people. Although human error still exists, machine learning is progressing at a rapid rate, and human error is continually decreasing. Hopefully, I will be able to delve deeper into this topic in Database Organization next semester!
I have attended the Machine Learning seminar by Vinesh Kannan, who is working in Google but has had an internship in Civic Digital Fellowship this summer. His talk was about how they used Machine Learning in finding certain quantity of “job names” for companies in each industry.
He first talked about how job names are categorized, and the early stages of their research. As long as the quantity was similar to the market demand, the program was successful. But then they noticed that some of the required jobs were “washed away.” The reason why these happened was that these jobs were containing words that exist in various industries, like data analyst and economic analyst. He said that they resolved the problem by commanding that each job that contain such words will be accepted as different jobs, so that they won’t disappear.
As a freshman year, these were a little hard for me to keep up while listening and actually understand what he meant. Most of the things I wrote were from what I understood from my discussion with some of the audience after the seminar.
Machine Learning seems to be an interesting topic. Until now, I saw it applied as a program that picks words and predicts based on them; both in this seminar and previously in the DemonHacks as the winner back then used ML to predict health problems based on text messages.
IIT has a ML club (ML@IIT), and I want to be active in that community.
I attended Vinesh Kannan’s talk on Machine learning at the Bureau of labor statistics. Going in to the lecture I was curious on what type of work he did as I didn’t have any clue of what the BLS would be working on in terms of machine learning. They were trying to be able to aid the process in assigning different jobs soc codes that represent specific categories, and to do this they implemented machine learning. I think he said that there were like 10 million records that they had in their data set, which is a mindblowing number to me.
Although some of the talk was out of my scope, it was interesting to see how difficult it is to train these models and the weird problems that arise. Example of one was the term analyst appearing in many different fields so it essentially was being ignored in the models predictions, which led to wrong classifications. I learned a-lot about this type of stuff, and hopefully it will come in handy during my own personal machine learning projects. The fellowship seems like a very cool opportunity that I would love to do down the line, since machine learning is fascinating to me and I like the focus the work has on using technology for good. Overall I was really glad I have joined the ml@iit club as this has been one of many talks I have attended that were really fun and a good use of time in between classes! I will definitely be going to more as well
I attended a seminar on biochemical engineering by a professor at northwestern university, I didn’t know what the professor was going to talk about before I went to this seminar, because I’m a computer science major. And I’m afraid I don’t understand it at all. But when I arrived, my worries disappeared. The professor is a very good speaker, he can give a vivid explanation of knowledge, so that everyone can understand and think about it.
He first introduced his team, which also included his students. It was an excellent team. They explored a lot and had a lot of academic achievements. And then he talks a lot about academics.
For example, catalysis in energy processes, food can change state through the environment(like crispy taco and soft taco), how to establish a well-understood base case, microbial electrolysis, etc. I’m interested in all of them.
The seminar made me want to attend all kinds of seminars. Because I am curious about different things, it can enrich my knowledge, so I will actively participate in the seminar of various subjects in the future.
I attended a seminar on november 20th presented by Professor Justin Notestein, from the Dept. of Chemical and Biomolecular Engineering at Northwestern University, about active sites and how to manipulate them. I felt kind of scared at first because it was my first event here and I didn’t really know how things were going to go down so I was not sure if I was really going or not, but last minute I made up my mind and decided to go. And afterall it turned out to be a good experience in the end. Through the event I learned that you can put a metal on top of a catalyst to either cover the active sites completely or just cover the sites in which the bondage is less selective (as they will bond to the metal), what enables scientists to manipulate those catalysts to only synthesise certain things. This activity helped me understand how computer science can be used in scenarios in which I never thought they could. For example in this case computer science can be used to help them study the active sites of the catalysts and their properties in a more detailed way. Even though I’m not a big chemistry guy, I enjoyed the experience therefore I probably would attend more activities like that. The only thing I would advise is to have some background on the subject as it can sometimes be hard to understand part of the vocabulary that is being used.
On November 21st, I attended the Machine Learning seminar by Vinesh Kannan about his experiences working as a data science fellow at the Bureau of Labor Statistics. I was nervous before going to the seminar because I was a afraid of not being able to comprehend what’s the main point the talker is trying to express. But after arriving, I think that most of my anxiety lessen and I was able to concentrate on the talk. His main point in the seminar is that how him and his team used Machine Learning in finding job names for requested companies from different fields.
He goes in deeper using Machine on how different job names are separated into different categories suitable for the different purposes of research. Something new I learned from the talk is that for his team to success on produces occupation and wage data used by policymakers, hiring staff, job seekers, and researchers across the country, the quantity of job names has to be the same to the market demand. He goes on to mention that some of the required jobs seems to be unavailable and the reason is that a lot of industries also have key words associated with these jobs, they solved it by separating jobs that have those kinds of words be different jobs.
Machine learning was very helpful and interesting. Though I cannot grasp 100% of the entire seminar, I came back with a good sense of how Machine Learning works. I would attend these activities again.
I attended the machine learning seminar on November 21st by Vinesh Kannan. His seminar focused on his experience withing as a fellow at the Bureau of Labor Statistics specializing in data science. I was eager to go to the seminar because data science and machine learning are two things I’m very interested in and hope to study while completing my computer science degree. Vinesh’s talk was very professional and informative. He went into great detail about the type of work he did and how he completed it using advanced machine learning models and some of the largest labor-related data sets on the planet. I was especially intrigued by his explanation about how the Bureau of Labor Statistics categorized jobs at a very intricate level. Before attending this seminar, I wouldn’t have considered that such a massive amount of data needs to be collected on every working person and every job, and that it must be categorized in such a complex way. I also wouldn’t have guessed that machine learning as complex as Vinesh described would be implemented in order to do all of this work.
After attending this seminar, I’m still very interested in machine learning and data science, and applying for the data science fellowship with the Bureau of Labor Statistics is definitely something I plan on doing in the future.
I attended a seminar by Vinesh Kannan, who talked about Machine Learning. Before going to the seminar, I didn’t have much knowledge about Machine Learning. So this experience for me to learn something from Vinesh Kannan was an exciting experience. I will understand why most people will have been nervous when going into the seminar without any knownledge about Machine Learning. But me on the other hand I was nervous at first, which later turned into excitement because of how Vinesh Kannan described the field.
Vinesh Kannan explains what him and his team used Machine Learning to do. Kannan describes how his teams objective was to find certain jobs to a certain company. For the teams objective to be successful Kannan and his team had to separate the different jobs and those jobs had certain words that you can group them with similar jobs that have those same words. After the seminar was over my understanding of Machine Learning was much better than before the Seminar.
This Seminar was a great Seminar and I learned alot from Vinesh Kannan. I plan on going to more Seminars that are about different topics in the Computer Science field. I hope to use what ever I learned from the Seminars in the future.
I attended a seminar on november 21st presented by Plamen Petrov, Vice President, Artificial Intelligence and Exponential Technologies and Chief Data Officer at Anthem Inc., about Introducing AI into the digital-first healthcare delivery. When I was going to this event I felt interested and pretty excited about going to it, as the topic was really interesting and I was looking forward to learning more about that sometime this year. So I decided that it was a good time to start getting more involved with the topic. I learned that Thirty-five percent of U.S. adults have gone online to self-diagnose a medical condition that they or someone they knew had, and that multiple times they find things that are not accurate to what’s going on or find “false-answers” to their problems. So through this talk I learned that AI can be introduced to this, in a way in which they would learn the types of problems people are searching, the solution they found and if it worked or not, this way the AI would learn what problems people are facing and accurately deliver a solution to them. I would definitely attend another activity like this, as it was a really good experience and it got me more connected to the school and to computer science, especially if it is a topic that sparks interest in me or it is something that relates to something I want to learn or know about.
I attended the Machine Learning seminar hosted by Vinesh Kannan, an IIT Computer Science graduate on the 21st of November, who currently works at Google but shares his experiences working as a data science fellow at the Bureau of Labor Statistics (BLS). Before going to the seminar, I was nervous because I don’t have much experience in the CS-related field, and I won’t be able to comprehend and stay on the same page as the speaker. Still, at the same time, I was excited to learn more about his experience in data science and Machine Learning. As the seminar started, I became more and more comfortable with understanding the speaker’s thoughts and what he was trying to convey. His primary message in the seminar was about how he and his fellow team members used Machine Learning in finding job names and positions for different companies in different fields. He went deeper into his experience of using Machines to successfully sort the various categories and subcategories for different research purposes. I learned that Vinesh and his team used data on occupation and wages made by researchers, future employees, and employers. He told me about how data was used to measure the number of jobs and also how Machines were used to help the employers get their best matches for candidates for their positions. Even though I didn’t understand every aspect of what Vinesh was trying to convey, I learned about how Machine Learning can be used in the industrial side of things, and I would love to attend future events to learn more and expand my knowledge.
On November 21st, I attended a seminar presented by Mr. Plamen Petrov. Mr. Petrov is the Vice President, Artificial Intelligence and Exponential Technologies and Chief Data Officer of Anthem Inc. His seminar was about being a company that integrates Artificial Intelligence into healthcare. Before going to this event, I felt extremely excited about the concept as AI is a subject that I am super passionate about. Just the thought of integrating computing and technology into everyday life is mind blowing to me. Also, I was eager to learn more about the concept as I knew there was plenty for me left to learn. I learned that 35% of people in the Unites States research their symptoms online only to find the wrong diagnosis and a treatment that is not accurate for their true condition. I learned that through AI, people can connect their symptoms to those who have already been treated and get the proper diagnosis and treatment for their needs. I found that this way would be a lot more effective and seamless rather than having to deal with the wrong diagnosis more than 1/3 of the time. I am very happy to have gone to seminar like this and I would definitely attend another activity like this in the future. I felt as if this activity brought me closer to knowing more about my major and feel as if it is a catalyst for me to learn and know more things about the concept.
Before participating in the machine learning seminar activity, I didn’t really feel much other than getting ready to learn new things regarding data science, machine learning and much more. I felt quite confident that I would walk out with a series of loaded information.
Throughout this event I learned a lot about the internship known as the 2020 Civil Digital Fellowship program including the projects they attempted including data science, product management, and the software they used. I also learned that these students have made groundbreaking progress in terms of where they currently are. They pursued top graduate schools including many very prominent companies for work.
I will definitely attend an activity like this again. During my first semester, I haven’t really attended many talks from research professors or even speakers from different universities. However, after listening to this one, I have discovered that these events are very well performed and will provide a ton of information regarding the major you have selected and will pursue. This knowledge that one learns will be important for future use, so I might as well continue to attend talks like this to broaden my knowledge in this certain topic, and continue to better myself in my selected field.
I attended the seminar on 11/21/2019 by Plamen Petrov. This talk largely focused on the usage of AI in healthcare systems as a preventative tool. Before this talk, I honestly had very little background understanding of healthcare applications as I had never used one in the past, although I did have an interest in the AI aspect. One thing that I learned from the talk is how much well protected any information is pertaining to someones health information. For example, Petrov’s application was forced to use simulated data in some cases because of the strictness of these rules. My learning of computer science has been expanded as I had a previous interest in AI, so learning some of the struggles of producing an AI has further sparked my interests in AI.
I would also definitely like to attend another talk by Petrov. I personally think that it is important for anyone going into any field to hear and learn from someone who has been successful in it. To me, Petrov is the shining example of someone who has found success in computer science and AI systems, and has made strides to further enrich others planing on joining the field through his companies multiple college outreach programs and research opportunities.
I have also went to the Seminar by Plamen Petrov on Thursday, 11/12.29. I rarely go to seminars, but after attending one, my thoughts about it has changed. Before, I view seminars as long boring lectures with lots of information to memorize. There are lots of great ideas that need to be shared. This seminar focuses on making a algorithm about health. Specifically, one that could diagnose a patient without a doctor. This is a work in progress, but I could see that potential it can grow into. It would be an app one can download. Then the patients list symptoms, problems, or take pictures. This is actually harder than it looks and brings a whole new set of problems in ways I never have thought of. First, the algorithm needs to be able to analyze text and pictures. Because of the hundreds of ways a person can describe a symptom along with the massive data that requires to be effective, it becomes a challenge.
I was better than I had expected. I was expecting to fall asleep, but I learned a lot about the problems of having a algorithm to interact with people. I enjoyed this seminar and would go to a similar one.
Before going to the machine learning seminar I did not really attend any other lectures. I was very skeptical and thought it would be very boring. I was also very excited to learn more about cs and cs oriented stuff because I became a CS major with no prior experience with it. Throughout the seminar, I learned a lot about how the speaker made connections with other companies and how they landed internships and future jobs. The main point that I learned from the seminar was the ways that Kannan and his team members had to collect data and use it to help sort different job types with different job positions. One thing that stood out to me was when he mentioned using names to associate them with different job titles. When the seminar ended I was still a little confused about some of the things he talked about. I thought this was a good experience I will hopefully be attending more talks in the future.
Before attending Vinesh Kannan’s Seminar, I did not really know how I felt about the seminar. I did not completely understand what a fellowship is and I did not really know what he was doing at the fellowship. Listening to his experience with the fellowship really peaked my interest however as he had the opportunity to attend a government job, which covered his flights, living, and most expenses he had to take upon himself and he also got paid for the work he did. Although it was only a couple of weeks, he put his full-time job on hold to experience this fellowship. Him pushing back his job for this experience really shows how much importance this fellowship had to him and as he stated during the seminar, he did not regret doing that. He learned many things and got to explore a part of the country he has never been in. It was also interesting to learn about the fellowship itself, as he got to work at the Bureau of Labor Statistics, which oversees all the data we look at when we search up salaries based on different criteria. He explained to us how this job works and how they have slowly introduced machine learning that outputs the 6-digit code they have for each different job that is in the market. This was always done by hand and introducing machine learning made everything a lot more efficient and quicker. Although there were still plenty of mistakes, the human input in sorting the data grew a lot smaller and the data could be processed much quicker. This activity helped me understand CS a lot more as it showed me a real-life job that deals with coding on a daily basis and it also introduced me into fellowships that I may want to explore later on in my college career. I would happily attend another event like this as it gives me insight into the CS world and how it all works.
I went to the Machine Learning seminar on November 21st. The seminar was presented by Vinesh Kannan, a graduate from IIT. Before attending the seminar, I was unfamiliar and curious about machine learning. I wanted to better understand why machine learning is an important part of computer science. I am a computer science major, but I still do not know what I want to specialize in. Vinesh Kannan talked about his worked experience for the Bureau of Labor Statistics. He used machine learning to categorize different types of job titles for companies. He also talked about how it could be problematic if two jobs have identical words in their names, so he talked about how his team was able to separate similar results into separate job titles. After the seminar ended, I felt like I have learned a lot more about machine learning. This talk also expanded my interests in computer science and could possibly lead me into a certain career in the future. I was undecided about what career I wanted to pursue in the computer science field, but machine learning sounds like it is an important job to work in. I think that I would attend another computer science related seminar again.
On November 21st I attended a Machine Learning seminar talk by Vinesh Kannan. Beforehand, I was eager to go because machine learning is one of my favorite topics in cs. His seminar focused on his experience within as a fellow at the Bureau of Labor Statistics specializing in data science. The talk was very informative and professional. I already know a good amount of behind how machine learning works but I don’t have much experience with seeing how it is implemented to a practical application. Seeing the problem presented and then an explanation of what was used for training data, how certain problems were solved was very interesting. For example, solving the prediction unrealistic jobs by creating context models to correlate job title and industry. His talk definitely improved my understanding of machine learning and its practical uses in the industry. It was also very interesting to learn how the data he produces is used by policy makers to make decisions.
After attending this seminar my interest in machine learning continues. Although, since this specific field is not what I am interested in, I don’t think I will pursue the data science fellowship with the Bureau of Labor Statistics
On November 21st, I attended a talk on the current condition of AI’s reach into the healthcare field, hosted by Vice President Plamen Petrov of Anthem Inc. I was mildly apprehensive, even if a but curious, expecting another future-looking series of aspirations on how AI could improve the field “one day” but what I got was a reasonable, non-theoretical application of the technology.
Because of the current state of the Amercian Healthcare System and many Americans inability to afford decent coverage, Anthem Inc. seeks to build a simpler, more affordable alternative through the device that over 80% of Americans current possess: a smartphone. Their app, CareSpree, mimics a typical doctor-patient consultation and compares the subjective analysis of the patient to “People Like Me”: their categorization database comparing people of similar symptoms, backgrounds and ethnicity or race.
If nothing else, it taught me about the reaches of AI pattern recognition and its applications outside of facial recognition technology. Despite Computer Science being my major of choice, I can’t say I get much information about the current state of the AI field outside of wildly exaggerated articles. It’s already being put to test in imaging and video analysis to work towards dermatology and cardiology studies.
On Thursday, October 17, I attended the talk “Differential Privacy, Adaptive Data Analysis, and Free Speedups via Sampling” by mathematics associate professor Lev Reyzin from the University of Illinois at Chicago. When I saw the emails about the talk, I read the abstract and saw that the talk promised to show a “recently-discovered beautiful connection between differential private mechanisms and preventing overfitting in machine learning.” This immediately made me interested in attending it since I had some prior experience in machine learning from both a high school research project with computer vision (convolutional neural nets) and a high school CS class. Additionally, I had learned before about preventing overfitting in neural network-based models using a dropout layer and in decision tree-based models using pruning, so I was intrigued to find out what other methods there were to prevent overfitting in other machine learning models.
When I walked into the room for the talk, I felt instantly out of place. Most of the people in the room looked much older than me, and I assumed they were mostly graduate or doctoral students with specializations in AI. Nevertheless, I ignored the impostor syndrome and took a seat, opening up my notebook to jot down anything I could understand or find useful.
The talk was pretty difficult to understand mainly due to the mathematical notation he was using, which involved a lot of probability theory. He talked about switching from a static-based machine learning model to an adaptive-based one, and he went into depth about different types of queries (statistical, counting, etc.) that could be made to a model. I couldn’t quite grasp all of this, and to be honest, if you asked me about any of this, I probably wouldn’t be able to answer you with 100% confidence. However, what I took away from his talk (and what I found most interesting, coincidentally), was that you could prevent overfitting by applying a slight Laplace distribution error to the training dataset and subsampling it.
As someone who likes to dabble in a variety of CS-related fields, I’d be lying if I said I didn’t want to go to other CS talks after this slightly difficult experience. However, what I do know now is that I should probably choose one that is more suited towards my CS experience level! Perhaps a more intermediate talk on AI/ML would have been more productive for me, rather than this graduate-level one.
On November 21st, I attended a Machine Learning seminar talk by Vinesh Kannan at IIt. I was excited to go to this seminar because machine learning a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention. This intrigues me because I want to be able to learn more about machine learning so that one day I will be able to do it too. Vinesh Kannan was a graduate of IIT, which made me even more interested in his talk since I will also be a graduate from IIT one day. He focused his talk on how he and his colleagues use machine learning in finding different jobs and positions for different companies in all kinds of fields. He emphasized a lot how a lot of jobs are always done by hand and using machine learning it made everything a lot easier. He inspired me into learning about machine learning, he was also a great public speaker that inspired me not only on machine learning but also in helping communities around me. I will be attending a lot more seminars to learn more about how I can use computer science to make the world a better place.
I attended Vinesh Kannan’s seminar on Machine Learning. “Machine learning is the scientific study of algorithms and statistical models that computer systems use to perform a specific task without using explicit instructions” – Wikipidea. Most of what I knew about Machine Learning came from what I googled ten minutes before the seminar. To be honest, I was expecting it to just be another boring, mandated event we had to attend for this class. But that was not the case; after attending the event, I learned some new and intriguing information, and am even a little bit more fascinated by Machine Learning now. I was particularly interested in Kannan’s discussion about his experience at the Bureau of Labor Statistics, and how they would use a six digit code to generate information about different jobs, and how they logged all information about job salaries in their database as well. Another interesting topic Kannan brought up is how Machine learning will automate jobs that most people thought could only be done by people. Although human error still exists, machine learning is progressing at a rapid rate, and human error is continually decreasing. Hopefully, I will be able to delve deeper into this topic in Database Organization next semester!