Jay Shah Podcast

Emma is a final-year medical student at the University of Cambridge and also pursuing her Ph.D. in Machine Learning. With her knowledge of clinical decision-making, she is working on research projects that leverage machine-learning techniques to improve clinical workflow. She will be taking her role as an academic doctor post her graduation. Time stamps of the conversation00:00:00 Introduction00:02:08 From clinical science to learning AI00:13:15 Learning the basics of Artificial Intelligence00:20:12 Promise of AI in medicine00:30:13 Do we really need interpretable AI models for clinical decision-making? 00:38:47 Using AI for more clinically-useful problems00:50:55 Facilitating interdisciplinary efforts00:54:06 Predicting length of stay in ICUs using convolutional neural networks01:03:04 AI for improving clinical workflows and biomarker discovery   01:07:55 Clustering disease trajectories in mechanically ventilated patients using machine learning01:16:37 ChatGPT for medical research or clinical decision making01:25:21 Quality over quantity of AI works published nowadays01:31:07 Advice to researchersEmma's Homepage: https://emmarocheteau.com/LinkedIn: https://www.linkedin.com/in/emma-rocheteau-125384132/Also check-out these talks on all available podcast platforms: https://jayshah.buzzsprout.comAbout the Host:Jay is a Ph.D. student at Arizona State University.Linkedin: https://www.linkedin.com/in/shahjay22/Twitter: https://twitter.com/jaygshah22Homepage: https://www.public.asu.edu/~jgshah1/ for any queries.Stay tuned for upcoming webinars!***Disclaimer: The information contained in this video represents the views and opinions of the speaker and does not necessarily represent the views or opinions of any institution. It does not constitute an endorsement by any Institution or its affiliates of such video content.***Checkout these Podcasts on YouTube: https://www.youtube.com/c/JayShahmlAbout the author: https://www.public.asu.edu/~jgshah1/

Understanding why and how transformers are so efficient in large language models nowadays such as #chatgpt and more.Watch the full podcast with Dr. Surbhi Goel here: https://youtu.be/stB0cY_fffoFind Dr. Goel on social media Website: https://www.surbhigoel.com/Linkedin: https://www.linkedin.com/in/surbhi-goel-5455b25aTwitter: https://twitter.com/surbhigoel_?lang=enLearning Theory Alliance: https://let-all.com/index.htmlAbout the Host:Jay is a Ph.D. student at Arizona State University.Linkedin: https://www.linkedin.com/in/shahjay22/Twitter: https://twitter.com/jaygshah22Homepage: https://www.public.asu.edu/~jgshah1/ for any queries.Stay tuned for upcoming webinars!***Disclaimer: The information contained in this video represents the views and opinions of the speaker and does not necessarily represent the views or opinions of any institution. It does not constitute an endorsement by any Institution or its affiliates of such video content.***Checkout these Podcasts on YouTube: https://www.youtube.com/c/JayShahmlAbout the author: https://www.public.asu.edu/~jgshah1/

Anupam is the co-founder and President of TruEra and prior to that, he was a Professor at Carnegie Mellon University for 15 years. TruEra provides AI solutions that help enterprises use machine learning, improve and monitor model quality, and build trust. His research and other efforts are focused on privacy, fairness, and building trustworthy machine-learning models. He holds a Ph.D. in computer science from Stanford University and Bachelor’s degree in same from IIT Kharagpur in India.Time stamps of the conversation00:50 Introductions01:45 Background and TruEra05:30 Trustworthy AI11:55 Validating Large models in the real world 16:15 History of NLP and large language models29:25 Opportunities and challenges with ChatGPT36:52 Evaluating the reliability of ChatGPT39:10 Existing tools that aid explainability 43:12 AI trends to look for in 2023 More about Dr. DattaWebsite: https://www.andrew.cmu.edu/user/danupam/Linkedin: https://www.linkedin.com/in/anupamdattaResearch: https://scholar.google.com/citations?user=oK3QM1wAAAAJ&hl=enAbout TruEra: https://truera.com/About the Host:Jay is a Ph.D. student at Arizona State University.Linkedin: https://www.linkedin.com/in/shahjay22/Twitter: https://twitter.com/jaygshah22Homepage: https://www.public.asu.edu/~jgshah1/ for any queries.Stay tuned for upcoming webinars!***Disclaimer: The information contained in this video represents the views and opinions of the speaker and does not necessarily represent the views or opinions of any institution. It does not constitute an endorsement by any Institution or its affiliates of such video content.***Checkout these Podcasts on YouTube: https://www.youtube.com/c/JayShahmlAbout the author: https://www.public.asu.edu/~jgshah1/

Surbhi is an Assistant Professor at the University of Pennsylvania. She got her Ph.D. in Computer Science from UT Austin and prior to joining UPenn as an Assistant Professor, she was a postdoctoral researcher at Microsoft Research NYC in the Machine Learning group. She has research expertise in theoretical computer science & machine learning, with a particular focus on developing theoretical foundations for modern deep learning paradigms. She also is a part of building the Learning Theory Alliance community that organizes and conducts several events useful for researchers and students in their careers. Time stamps of the conversation00:00:54 Introduction00:01:54 Background and research interests00:05:03 Interest in Machine Learning Theory00:13:02 Understanding how deep learning works00:16:30 Transformer architecture00:25:40 Scale of data and big models00:31:28 Reasoning in deep learning 00:38:52 Theoretical perspective on AGI, consciousness, and sentience in AI00:46:00 Remaining updated to the latest research00:53:38 Should one do a Ph.D.? 00:57:45 Is a Ph.D. mandatory for machine learning industry positions?01:01:38 What makes a good research thesis?01:05:30 Some best practices in research01:12:20 Learning Theory Alliance Group01:14:25 Job interviews in academia for researchers01:20:00 Advice to young researchers and students01:25:02 Decision to become a ProfessorFind Dr. Goel on social media Website: https://www.surbhigoel.com/Linkedin: https://www.linkedin.com/in/surbhi-goel-5455b25aTwitter: https://twitter.com/surbhigoel_?lang=enLearning Theory Alliance: https://let-all.com/index.htmlAbout the Host:Jay is a Ph.D. student at Arizona State University.Linkedin: https://www.linkedin.com/in/shahjay22/Twitter: https://twitter.com/jaygshah22Homepage: https://www.public.asu.edu/~jgshah1/ for any queries.Stay tuned for upcoming webinars!***Disclaimer: The information contained in this video represents the views and opinions of the speaker and does not necessarily represent the views or opinions of any institution. It does not constitute an endorsement by any Institution or its affiliates of such video content.***Checkout these Podcasts on YouTube: https://www.youtube.com/c/JayShahmlAbout the author: https://www.public.asu.edu/~jgshah1/

Sebastian Raschka​ is the lead AI educator at GridAI. He is the author of the book "Machine Learning with PyTorch and Scikit Learn" and also a few other books that cover the fundamentals of #machinelearning and #deeplearning techniques and implementing them with Python. He is also an Assistant Professor of Statistics at the University of Wisconsin-Madison and has been actively involved in making ML more accessible to beginners through his blogs, video tutorials, tweets and of course his books. He also holds a doctorate in Computational and Quantitative Biology from Michigan State University.Time Stamps of the Podcast00:00:00 Introductions00:02:40 Entry point in AI/ML that made you interested in it00:05:30 How did you go about learning the basics and implementation of various methods?00:11:45 What makes Python ideal for learning Machine Learning recently?00:21:54 What is your book about and who is this for?00:33:55 What goes into writing a good technical book?00:40:50 Applying ML to toy datasets vs real-world research problems00:47:40 Choosing b/w machine learning methods & deep learning methods00:56:22 Large models vs architecture efficient models 01:01:25 Interpretability & Explainability in AI01:08:45 Insights for people interested in machine learning research, academia or PhD01:14:17 Keeping up with research in deep learningSebastian's homepage: https://sebastianraschka.com/Twitter: https://mobile.twitter.com/rasbtLinkedIn: https://www.linkedin.com/in/sebastianraschka/His book: https://www.amazon.com/Machine-Learning-PyTorch-Scikit-Learn-scikit-learn-ebook-dp-B09NW48MR1/dp/B09NW48MR1/Video Tutorials:  @SebastianRaschka  About the Host:Jay is a Ph.D. student at Arizona State University.Linkedin: https://www.linkedin.com/in/shahjay22/Twitter: https://twitter.com/jaygshah22Reach out to https://www.public.asu.edu/~jgshah1/ for any queries.Stay tuned for upcoming webinars!***Disclaimer: The information contained in this video represents the views and opinions of the speaker and does not necessarily represent the views or opinions of any institution. It does not constitute an endorsement by any Institution or its affiliates of such video content.***Checkout these Podcasts on YouTube: https://www.youtube.com/c/JayShahmlAbout the author: https://www.public.asu.edu/~jgshah1/

Dr. Matthew Lungren is currently the Chief Medical Information Officer at Nuance Communications - Microsoft company, and also holds part-time appointments with the University of California San Francisco as an Associate Clinical Professor and also as adjunct faculty at Stanford and Duke University. He is a radiologist by training and has led and contributed to multiple projects that use AI and deep learning for medical imaging and precision medicine. Time stamps from the conversation00:00:55 Introduction00:01:46 Role as a Chief Medical Information Officer 00:05:25  Leading research projects in the industry00:08:45 Is AI ready for primetime use cases in the real world?00:12:40 Regulations on AI systems in healthcare00:17:25 Interpretability vs a robust validation framework00:25:22 Promising directions to mitigate data issues in medical research00:32:24 Stable diffusion models 00:34:06 Making datasets public00:39:00 Vision transformers for multi-modal models00:44:35 Biomarker discovery00:48:20 Sentiment of AI in medicine 00:53:26 Bridging the communication gap between computer scientists and medical experts01:01:42 Advice to young researchers from medical and engineering schoolsFind Dr. Lungren on social media Twitter: https://twitter.com/mattlungrenmdLinkedIn: https://www.linkedin.com/in/mattlungrenmd/About the Host:Jay is a Ph.D. student at Arizona State University.Linkedin: https://www.linkedin.com/in/shahjay22/Twitter: https://twitter.com/jaygshah22Homepage: https://www.public.asu.edu/~jgshah1/ for any queries.Stay tuned for upcoming webinars!***Disclaimer: The information contained in this video represents the views and opinions of the speaker and does not necessarily represent the views or opinions of any institution. It does not constitute an endorsement by any Institution or its affiliates of such video content.***Checkout these Podcasts on YouTube: https://www.youtube.com/c/JayShahmlAbout the author: https://www.public.asu.edu/~jgshah1/

Dr. Charles Fisher is the CEO and Founder of Unlearn(dot)AI which helps in faster drug development and efficient clinical trials. This year they also raised a series B funding of 50 million dollars. Charles holds a Ph.D. in biophysics from Harvard University and prior to founding Unlearn, he did his Postdoctorate at Boston University, followed by being a principal scientist at Pfizer and a machine learning engineer at a virtual reality company in silicon valley. Time stamps of the conversation00:00:30 Introduction00:01:16 What got you into Machine Learning?00:04:10 Learning the basics and implementation00:07:55 Digital twins for clinical trials and drug development00:13:06 Patient heterogeneity in medical research00:16:05 Error quantification of models00:17:17 ML models for drug development00:22:45 Adoption of AI in medical applications00:25:35 Building trust in AI systems 00:35:10 How to show AI models are safe in the real world?00:38:38 Moving from academia to industry to entrepreneurship00:45:08 Research projects in startups vs academia vs big companies00:53:12 Routine as a CEO00:57:50 Is a Ph.D. necessary for a research career in the industry?01:01:20 Taking inspiration from biology to improve machine learning01:05:25 Advice to young peopleAbout Charles:LinkedIn: https://www.linkedin.com/in/drckf/More about Unlearn: https://www.unlearn.ai/About the Host:Jay is a Ph.D. student at Arizona State University.Linkedin: https://www.linkedin.com/in/shahjay22/Twitter: https://twitter.com/jaygshah22Homepage: https://www.public.asu.edu/~jgshah1/ for any queries.Stay tuned for upcoming webinars!***Disclaimer: The information contained in this video represents the views and opinions of the speaker and does not necessarily represent the views or opinions of any institution. It does not constitute an endorsement by any Institution or its affiliates of such video content.***Checkout these Podcasts on YouTube: https://www.youtube.com/c/JayShahmlAbout the author: https://www.public.asu.edu/~jgshah1/

Mina Ghashami is an Applied Scientist in the Alexa Video team at Amazon Science alongside being a lecturer at Stanford University. Prior to joining Amazon, she was a Research Scientist at Visa Research working on recommendation systems built on transactions from users and a few other projects. She completed her Ph.D. in Computer Science from the University of Utah followed by a PostDoctoral position at Rutgers University. At Amazon, she is mainly focused on Video-based ranking recommendation systems, something we talk about in detail in this conversation. Time stamps of the conversation00:00:50 Introductions00:01:40 Alexa Video - Ranking and Recommendation research00:05:25 Feature engineering for recommendation systems00:08:30  Ground truth for training recommendation systems00:12:46 What does an Applied Scientist do? (at Amazon)00:19:17 What got you into AI? And specifically recommendation systems00:24:30 Matrix approximation00:27:15 Challenges in recommendation research00:32:00 What's more interesting, theoretical or applied side of research?00:37:10 Over parametrization vs generalizability 00:39:55 Managing academic and industry positions at the same time00:46:26 Should one do a Ph.D. for research roles in the industry?00:50:00 Skills learned while pursuing a PhD00:54:22 Deciding industry vs academia00:56:20 Coping up with research in deep learning01:02:14 What makes a good research dissertation?01:04:16 Advice to young students navigating their interest in machine learningTo learn more about Mina:Homepage: https://mina-ghashami.github.io/Linkedin: https://www.linkedin.com/in/minaghashamiResearch: https://scholar.google.com/citations?user=msJHsYcAAAAJ&hl=enAbout the Host:Jay is a Ph.D. student at Arizona State University.Linkedin: https://www.linkedin.com/in/shahjay22/Twitter: https://twitter.com/jaygshah22Homepage: https://www.public.asu.edu/~jgshah1/ for any queries.Stay tuned for upcoming webinars!***Disclaimer: The information contained in this video represents the views and opinions of the speaker and does not necessarily represent the views or opinions of any institution. It does not constitute an endorsement by any Institution or its affiliates of such video content.***Checkout these Podcasts on YouTube: https://www.youtube.com/c/JayShahmlAbout the author: https://www.public.asu.edu/~jgshah1/

Alberto Santamaria-Pang is a Principal Applied Data Scientist at Microsoft.  He did his Ph.D. in computer science from the University of Houston and has a long experience in research and development on various AI projects including but not limited to medical imaging and deep learning. Prior to Microsoft, he was a principal scientist at GE research. He has led many research projects in industry and also government-funded projects, a few of which we will be discussing today. Time stamps of conversations:00:00:37 Introduction00:01:25 Background before you got into the industry00:04:17 Interest in AI and Medical Imaging00:05:54 What does a Principal Scientist do?00:10:00 What drives research in industry? Product or Theoretical pursuit?00:11:35 Learning skills relevant to a principal scientist00:15:14 Principal Investigator vs Principal Scientist00:21:00 How do industry and academia collaborate on research projects?00:25:30 Promise & challenges of AI in medical research and applications00:31:53 What should explainable AI look like?00:38:35 Adoption of AI in medical research00:43:00 Is AI generalizable? 00:44:36 AI for biomarker discovery00:51:42 Are large models useful in AI & Med space00:58:00 Why is there a lack of datasets?01:01:02 Do you think AI is scary?01:04:00 Where do we need innovation in AI precisely?01:10:20 Getting inspiration from bio-research to improve algorithms01:13:19 AI and molecular pathology for cancer research00:20:30 Should one get a Ph.D.?01:27:38 Advice for young people About Alberto:His research works: https://scholar.google.com/citations?user=sVahJxsAAAAJ&hl=enLinkedIn: https://www.linkedin.com/in/alberto-santamariaAbout the Host:Jay is a Ph.D. student at Arizona State University.Linkedin: https://www.linkedin.com/in/shahjay22/Twitter: https://twitter.com/jaygshah22Homepage: https://www.public.asu.edu/~jgshah1/ for any queries.Stay tuned for upcoming webinars!***Disclaimer: The information contained in this video represents the views and opinions of the speaker and does not necessarily represent the views or opinions of any institution. It does not constitute an endorsement by any Institution or its affiliates of such video content.***Checkout these Podcasts on YouTube: https://www.youtube.com/c/JayShahmlAbout the author: https://www.public.asu.edu/~jgshah1/

Are large language models really sentient or conscious? What is explainability (XAI) and how can we create human-aware AI systems for collaborative tasks? Dr. Subbarao Kambhampati sheds some light on these topics, generating explanations for human-in-loop AI systems and understanding 'intelligence' in context to AI systems. He is a Prof of Computer Science at Arizona State University and director of the Yochan lab at ASU where his research focuses on decision-making and planning specifically in the context of human-aware AI systems. He has received multiple awards for his research contributions. He has also been named a fellow of AAAI, AAAS, and ACM and also a distinguished alumnus from the University of Maryland and also recently IIT Madras.Time stamps of conversations:00:00:40 Introduction00:01:32 What got you interested in AI?00:07:40 Definition of intelligence that is not related to human intelligence00:13:40 Sentience vs intelligence in modern AI systems00:24:06 Human aware AI systems for better collaboration00:31:25 Modern AI becoming natural science instead of an engineering task00:37:35 Understanding symbolic concepts to generate accurate explanations00:56:45 Need for explainability and where01:13:00 What motivates you for research, the application associated or theoretical pursuit?01:18:47 Research in academia vs industry01:24:38 DALL-E performance and critiques01:45:40 What makes for a good research thesis? 01:59:06 Different trajectories of a good CS PhD student02:03:42 Focusing on measures vs metrics 02:15:23 Advice to students on getting started with AIArticles referred in the conversationAI as Natural Science?: https://cacm.acm.org/blogs/blog-cacm/261732-ai-as-an-ersatz-natural-science/fulltextPolanyi's Revenge and AI's New Romance with Tacit Knowledge: https://cacm.acm.org/magazines/2021/2/250077-polanyis-revenge-and-ais-new-romance-with-tacit-knowledge/fulltextMore about Prof. RaoHomepage: https://rakaposhi.eas.asu.edu/Twitter: https://twitter.com/rao2zAbout the Host:Jay is a PhD student at Arizona State University.Linkedin: https://www.linkedin.com/in/shahjay22/Twitter: https://twitter.com/jaygshah22Homepage: https://www.public.asu.edu/~jgshah1/ for any queries.Stay tuned for upcoming webinars!***Disclaimer: The information contained in this video represents the views and opinions of the speaker and does not necessarily represent the views or opinions of any institution. It does not constitute an endorsement by any Institution or its affiliates of such video content.***Checkout these Podcasts on YouTube: https://www.youtube.com/c/JayShahmlAbout the author: https://www.public.asu.edu/~jgshah1/