Explore a comprehensive video explanation of a machine learning research paper on efficient and modular implicit differentiation. Delve into advanced topics like automatic differentiation of inner optimizations, meta-learning, optimization unrolling, and the implicit function theorem. Learn about a unified framework for implicit differentiation of optimization problems that combines autodiff benefits with efficiency and modularity. Discover how this approach can be applied to bi-level optimization problems and sensitivity analysis in molecular dynamics. Follow along with the detailed outline covering key concepts, mathematical foundations, and experimental results presented by the speaker.
Efficient and Modular Implicit Differentiation - Machine Learning Research Paper Explained
Overview
Syllabus
- Intro & Overview
- Automatic Differentiation of Inner Optimizations
- Example: Meta-Learning
- Unrolling Optimization
- Unified Framework Overview & Pseudocode
- Implicit Function Theorem
- More Technicalities
- Experiments
Taught by
Yannic Kilcher
Related Courses
-
Decision Awareness in Reinforcement Learning - End-to-End Optimization Approaches
-
IMAML- Meta-Learning with Implicit Gradients
-
Meta-Learning - Why It’s Hard and What We Can Do - Ke Li
-
OpenAI - Solving Rubik's Cube with a Robot Hand - RL Paper Explained
-
Gradients Are Not All You Need - Machine Learning Research Paper Explained
