Learning and Search
Misha discusses the interplay between learning and search in AI, highlighting how AlphaGo exemplifies the power of combining these elements. He notes that while language models excel in multitasking, they struggle with reliability due to error accumulation. The challenge remains to enhance the competence of general agents by incorporating effective planning and search strategies.In this clip
From this podcast
Training Data
Reflection AI’s Misha Laskin on the AlphaGo Moment for LLMs | Training Data
Related Questions