Memory and Learning
Roland discusses the intricate relationship between memory and learning, highlighting the concept of fast weights, where learning can occur at various time scales. He contrasts instantaneous fact learning with the slower synaptic adaptation seen in neural networks, raising questions about how these two forms of learning interact and influence each other.In this clip
From this podcast
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
What’s Next in LLM Reasoning? with Roland Memisevic - 646
Related Questions