Chapter
Learning-Based Systems vs Classic Optimal Control Systems
35:23 - 44:12 (08:49)
The difference between learning-based systems and classic optimal control systems is that the former's performance gets better with time but the question of explainability is closely tied to its performance.
Clips
The accumulation of experiences in data form to form deep representations starts to feel like knowledge as opposed to optimal control, making learning-based systems get better over time.
35:23 - 39:49 (04:26)
Summary
The accumulation of experiences in data form to form deep representations starts to feel like knowledge as opposed to optimal control, making learning-based systems get better over time. Deep learning has inspired thousands of researchers and has shown incredible results and breakthroughs.
ChapterLearning-Based Systems vs Classic Optimal Control Systems
Episode#108 – Sergey Levine: Robotics and Machine Learning
PodcastLex Fridman Podcast
The question of explainability in learning-based systems is complicated and closely tied to the system's performance.
39:49 - 42:37 (02:48)
Summary
The question of explainability in learning-based systems is complicated and closely tied to the system's performance. While expert systems can be easily explained through its reasoning process, interpretability and explainability in learning-based systems is somewhat lacking and requires further development.
ChapterLearning-Based Systems vs Classic Optimal Control Systems
Episode#108 – Sergey Levine: Robotics and Machine Learning
PodcastLex Fridman Podcast
Jacob explores the relationship between reinforcement learning and natural language processing, examining how policies can be structured with NLP for better outcomes in completing tasks.
42:37 - 44:12 (01:34)
Summary
Jacob explores the relationship between reinforcement learning and natural language processing, examining how policies can be structured with NLP for better outcomes in completing tasks.