"Mathematical Foundations of GenAI course lectures by IIT Madras"

View profile for Prathosh AP, PhD

Faculty member at IISc || Faculty member at IITD

Hello All, Here is a BIG announcement - TLDR - Video lectures of the course on Mathematical Foundations of GenAI. UPDATE 2 - The lecture on the state space models (SSMs) is out. All the videos of the course that I taught as a part of Indian Institute of Technology, Madras BS program, on *Mathematical Foundations of Deep Generative Modelling* are out in public. Here is the link to the playlist - https://lnkd.in/gZ-7ViTH UPDATE - Here is the link for the assignments - https://lnkd.in/gUwhC9AF The GitHub link for the assignments - https://lnkd.in/gxuZ_AKv I have carefully designed this course to include mathematical underpinnings of several generative models such as GANs, VAEs, Diffusion Models, Auto Regressive Models, Language Models, RLHF (PPO, DPO etc.) and more. Given the mathematical nature of the course, I have resorted to the old school, Chalk-n-talk style where I write out each and every equation with proofs. (The link to the lecture notes is present in the comment section). In addition, my student has taken several tutorials where every concept taught is implemented in Pytorch from scratch. While the internet is flooded with courses on GenAI, this type of content is very seldom to be found which motivated me to put it out. This is not a quick-byte or GenAI-in-X-days stuff, but content for those who care about the inner workings of the generative models. I hope that this would be of value to a few and welcome any feedback. #GenAI #lectures #iisc #iitm Indian Institute of Science (IISc) Department of ECE, IISc

Prathosh AP, PhD

Faculty member at IISc || Faculty member at IITD

2mo
Prathosh AP, PhD

Faculty member at IISc || Faculty member at IITD

1mo

Since a lot of people asked about a lecture on State space models (SSMs), I have recorded and uploaded it here - https://youtu.be/A7iAmVr0QE4?si=7p_VUHInUmlVloAl

Ganesh Paramasivam

Project Manager, AI Center of Excellence in Health, MAHE. Additional Professor, Cardiology, Kasturba Medical College, Manipal.

1mo

I am currently enrolled in this course as part of MTech (AI/ML) and loving it (PS: This is not an easy subject though). Prathosh AP, PhD sir, you made it easier through your lectures putting together great (and unique) material from diverse resources. The accompanying tutorials on pytorch implementation by Chandan are great too. I believe this could be written up as a book too. Thank you so much for making these mathematical concepts accessible to so many people, Prof Prathosh AP, PhD. 💐

Kumar Kaustubh

Technical Lead at HCLTech Engineering and R&D Services (ERS) AIx, Capability Unit | Speech and Natural Language Processing | Multimodal AI | IIT Dharwad

2mo

This is a huge contribution!! Thank you so much, sir! Honestly, I wish the chalk-n-talk were on a real blackboard with a moving camera 🥹

sir if possible can u also cover flow matching stuff

Rajesh Berepalli

Data Scientist @ ExxonMobil

2mo

Prathosh AP, PhD , this resource is outstanding. If you could complement it with assignments, it would become a truly comprehensive gateway for beginners.

Nabojwal Acharjee

PhD Candidate in Medical Image Analysis

2mo

Thank you Prathosh AP, PhD Sir, for this wonderful series! Sir, I have a small query regarding the SSMs -- Could you kindly clarify whether this topic is included in the playlist?

Like
Reply
Ramakrishnan A G

*Education *Transformation *Innovation *Holistic health *Pranayama

2mo

Great service, Prathosh!

Lalit Kumar

Veteran (Indian Air Force)

2mo

Sir, I am currently taking this course, and I want to thank you for the amazing work! Your style of teaching is truly unique and refreshing, the way you build intuition through mathematics and make even heavy derivations feel smooth and approachable is remarkable. Congratulations on putting this content out for everyone! PS- Do you also have similar material on Foundations of Machine Learning? I’d love to explore that as well. Best wishes and regards

See more comments

To view or add a comment, sign in

Explore content categories