Ian Osband
TalkRL: The Reinforcement Learning Podcast
Efficiency in Training Models with Bayesian Inference
Training models using an ensemble approach raises concerns about inefficiency and wastage of resources since each model is trained from scratch without reusing any learnings. Bayesian deep learning faces challenges due to the complexity of neural networks, creating difficulties in understanding the model's functioning and making Bayesian inference over these complex models less efficient, especially when dealing with numerous weights.
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.
