Jay Shah Podcast Jay Shah

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 Podcast
00:00:00 Introductions
00:02:40 Entry point in AI/ML that made you interested in it
00:05:30 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 problems
00:47:40 Choosing b/w machine learning methods & deep learning methods
00:56:22 Large models vs architecture efficient models
01:01:25 Interpretability & Explainability in AI
01:08:45 Insights for people interested in machine learning research, academia or PhD
01:14:17 Keeping up with research in deep learning

Sebastian's homepage: https://sebastianraschka.com/
Twitter: https://mobile.twitter.com/rasbt
LinkedIn: 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/jaygshah22

Reach 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/JayShahml
About 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 conversation
00:00:55 Introduction
00:01:46 Role as a Chief Medical Information Officer 
00:05:25  Leading research projects in the industry
00:08:45 Is AI ready for primetime use cases in the real world?
00:12:40 Regulations on AI systems in healthcare
00:17:25 Interpretability vs a robust validation framework
00:25:22 Promising directions to mitigate data issues in medical research
00:32:24 Stable diffusion models 
00:34:06 Making datasets public
00:39:00 Vision transformers for multi-modal models
00:44:35 Biomarker discovery
00:48:20 Sentiment of AI in medicine 
00:53:26 Bridging the communication gap between computer scientists and medical experts
01:01:42 Advice to young researchers from medical and engineering schools

Find Dr. Lungren on social media 
Twitter: https://twitter.com/mattlungrenmd
LinkedIn: 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/jaygshah22
Homepage: 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/JayShahml
About 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 conversation
00:00:30 Introduction
00:01:16 What got you into Machine Learning?
00:04:10 Learning the basics and implementation
00:07:55 Digital twins for clinical trials and drug development
00:13:06 Patient heterogeneity in medical research
00:16:05 Error quantification of models
00:17:17 ML models for drug development
00:22:45 Adoption of AI in medical applications
00:25:35 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 entrepreneurship
00:45:08 Research projects in startups vs academia vs big companies
00:53:12 Routine as a CEO
00:57:50 Is a Ph.D. necessary for a research career in the industry?
01:01:20 Taking inspiration from biology to improve machine learning
01:05:25 Advice to young people

About 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/jaygshah22
Homepage: 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/JayShahml
About 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 conversation
00:00:50 Introductions
00:01:40 Alexa Video - Ranking and Recommendation research
00:05:25 Feature engineering for recommendation systems
00:08:30  Ground truth for training recommendation systems
00:12:46 What does an Applied Scientist do? (at Amazon)
00:19:17 What got you into AI? And specifically recommendation systems
00:24:30 Matrix approximation
00:27:15 Challenges in recommendation research
00:32:00 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 time
00:46:26 Should one do a Ph.D. for research roles in the industry?
00:50:00 Skills learned while pursuing a PhD
00:54:22 Deciding industry vs academia
00:56:20 Coping up with research in deep learning
01:02:14 What makes a good research dissertation?
01:04:16 Advice to young students navigating their interest in machine learning

To learn more about Mina:
Homepage: https://mina-ghashami.github.io/
Linkedin: https://www.linkedin.com/in/minaghashami
Research: https://scholar.google.com/citations?user=msJHsYcAAAAJ&hl=en

About the Host:
Jay is a Ph.D. student at Arizona State University.
Linkedin: https://www.linkedin.com/in/shahjay22/
Twitter: https://twitter.com/jaygshah22
Homepage: 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/JayShahml
About 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 Introduction
00:01:25 Background before you got into the industry
00:04:17 Interest in AI and Medical Imaging
00:05:54 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 scientist
00:15:14 Principal Investigator vs Principal Scientist
00:21:00 How do industry and academia collaborate on research projects?
00:25:30 Promise & challenges of AI in medical research and applications
00:31:53 What should explainable AI look like?
00:38:35 Adoption of AI in medical research
00:43:00 Is AI generalizable? 
00:44:36 AI for biomarker discovery
00:51:42 Are large models useful in AI & Med space
00:58:00 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 algorithms
01:13:19 AI and molecular pathology for cancer research
00:20:30 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=en
LinkedIn: https://www.linkedin.com/in/alberto-santamaria

About the Host:
Jay is a Ph.D. student at Arizona State University.
Linkedin: https://www.linkedin.com/in/shahjay22/
Twitter: https://twitter.com/jaygshah22
Homepage: 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/JayShahml
About 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 Introduction
00:01:32 What got you interested in AI?
00:07:40 Definition of intelligence that is not related to human intelligence
00:13:40 Sentience vs intelligence in modern AI systems
00:24:06 Human aware AI systems for better collaboration
00:31:25 Modern AI becoming natural science instead of an engineering task
00:37:35 Understanding symbolic concepts to generate accurate explanations
00:56:45 Need for explainability and where
01:13:00 What motivates you for research, the application associated or theoretical pursuit?
01:18:47 Research in academia vs industry
01:24:38 DALL-E performance and critiques
01:45:40 What makes for a good research thesis? 
01:59:06 Different trajectories of a good CS PhD student
02:03:42 Focusing on measures vs metrics 
02:15:23 Advice to students on getting started with AI

Articles referred in the conversation
AI as Natural Science?: https://cacm.acm.org/blogs/blog-cacm/261732-ai-as-an-ersatz-natural-science/fulltext
Polanyi'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/fulltext

More about Prof. Rao
Homepage: https://rakaposhi.eas.asu.edu/
Twitter: https://twitter.com/rao2z

About the Host:
Jay is a PhD student at Arizona State University.
Linkedin: https://www.linkedin.com/in/shahjay22/
Twitter: https://twitter.com/jaygshah22
Homepage: 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/JayShahml
About the author: https://www.public.asu.edu/~jgshah1/