Machine Learning Street Talk (MLST)

ICLR 2020: Yann LeCun and Energy-Based Models

May 19, 2020

Yann LeCun, a pioneer in machine learning and AI, discusses the latest in self-supervised learning and energy-based models (EBMs). He compares how humans and machines learn concepts, advocating for methods that mimic human cognition. The conversation dives into EBMs' applications in optimizing labels and addresses challenges in traditional models. LeCun also explores the potential of self-supervised learning techniques for enhancing AI capabilities, such as in natural language processing and image recognition.

02:12:11

Creator website

Episode guests

Yann LeCun

AI Chapters

Episode notes

Intro

3min

Energy-Based Models in Machine Learning

17min

Self-Supervised Learning and Human Cognition

10min

Exploring Self-Supervised Learning and Multimodal Predictions

2min

Understanding Energy-Based Models

10min

Understanding Deterministic Models and Latent Variables in Variational Autoencoders

2min

Understanding Energy-Based Models and Latent Variables

15min

Exploring Manifolds and Machine Learning

10min

Challenges in Probabilistic Model Normalization

2min

10.

Exploring Energy-Based Models in NLP

16min

11.

Exploring Siamese Networks and Energy Functions in Image Recognition

2min

12.

Navigating Visual and Linguistic Complexity

12min

13.

Innovations in GAN Training

10min

14.

Reinforcement Learning vs. Model Predictive Control

4min

15.

The Challenges and Future of Self-Supervised Learning

2min

16.

Exploring the Nature of Intelligence and IQ

2min

17.

Exploring Intelligence: Constraints and Complexities

2min

18.

Exploring Energy-Based Models in Machine Learning

12min

This week Connor Shorten, Yannic Kilcher and Tim Scarfe reacted to Yann LeCun's keynote speech at this year's ICLR conference which just passed. ICLR is the number two ML conference and was completely open this year, with all the sessions publicly accessible via the internet. Yann spent most of his talk speaking about self-supervised learning, Energy-based models (EBMs) and manifold learning. Don't worry if you hadn't heard of EBMs before, neither had we!

Thanks for watching! Please Subscribe!

Paper Links:

ICLR 2020 Keynote Talk: https://iclr.cc/virtual_2020/speaker_7.html

A Tutorial on Energy-Based Learning: http://yann.lecun.com/exdb/publis/pdf/lecun-06.pdf

Concept Learning with Energy-Based Models (Yannic's Explanation): https://www.youtube.com/watch?v=Cs_j-oNwGgg

Concept Learning with Energy-Based Models (Paper): https://arxiv.org/pdf/1811.02486.pdf

Concept Learning with Energy-Based Models (OpenAI Blog Post): https://openai.com/blog/learning-concepts-with-energy-functions/

#deeplearning #machinelearning #iclr #iclr2020 #yannlecun

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.