Spatial Awareness Challenges
Wah discusses the complexities of training algorithms to navigate real-world environments, highlighting the challenges of transferring knowledge from virtual to physical spaces. The conversation delves into the limitations of deep reinforcement learning, particularly around sample efficiency and generalization, explaining why many applications, like delivery robots and self-driving cars, are not reliant on reinforcement learning at this stage.In this clip
From this podcast
Super Data Science: ML & AI Podcast with Jon Krohn
SDS 551: Deep Reinforcement Learning — with Wah Loon Keng
Related Questions