Ian Osband
TalkRL: The Reinforcement Learning Podcast
Distinguishing between Marginal and Joint Predictions
Machine learning often focuses on single input, single output predictions, known as marginal predictions. However, joint predictions involve multiple inputs and outputs, challenging the traditional one-to-one prediction model. Joint prediction highlights the interdependence or independence of predictions, emphasizing the importance of understanding the intricate relationships between multiple predictions. This concept aligns with information theory's mutual information, which captures the essence of interconnections within joint predictions.
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.
