
Teaching Large Language Models to Reason with Reinforcement Learning with Alex Havrilla - #680
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
Balancing Theoretical and Applied Aspects of Research in LLMs
The research broadly focuses on dividing intentions into two parts: theoretical aspects like neural network learning theory to make statements about generalization error and network size, and applied aspects like RL fine tuning for large scale experiments to improve reasoning capabilities of language models. The research involves automated feedback without human supervision in the training loop.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.