Learning Algorithms Unveiled
Craig explores the nuances of back relaxation versus greedy bottom-up learning, revealing that both approaches yield similar results. The conversation shifts to the importance of unsupervised pre-training in modern neural networks like BERT and GPT-3, emphasizing its resurgence in the face of larger datasets. Insights into reinforcement learning and its connection to brain function further enrich the discussion, highlighting the evolving landscape of computational neuroscience.In this clip
From this podcast
Eye on AI
Geoffrey Everest Hinton reviews his work in 2020 and talks about what he sees on the AI horizon
Related Questions
Is the predictive part of our brain bottom-up or top-down as discussed in the episode Perception: Chaos and Order | Dr. Karl Friston | EP 298 and the clip Brain Hierarchical Inference?
Is the predictive part of our brain bottom-up or top-down in the episode Perception: Chaos and Order | Dr. Karl Friston | EP 298 and the clip Brain Hierarchical Inference?