Stochastic Flows and Geometric Optimization on the Orthogonal Group

Abstract

We present a new class of stochastic, geometrically-driven optimization algorithms on the orthogonal group $O(d)$ and naturally reductive homogeneous manifolds obtained from the action of the rotation group $SO(d)$. We theoretically and experimentally demonstrate that our methods can be applied in various fields of machine learning including deep, convolutional and recurrent neural networks, reinforcement learning, normalizing flows and metric learning. We show an intriguing connection between efficient stochastic optimization on the orthogonal group and graph theory (e.g. matching problem, partition functions over graphs, graph-coloring). We leverage the theory of Lie groups and provide theoretical results for the designed class of algorithms. We demonstrate broad applicability of our methods by showing strong performance on the seemingly unrelated tasks of learning world models to obtain stable policies for the most difficult $\mathrm{Humanoid}$ agent from $\mathrm{OpenAI}$ $\mathrm{Gym}$ and improving convolutional neural networks.

References in corpus (7)

• Evolution Strategies as a Scalable Alternative to Reinforcement Learning
• Recurrent World Models Facilitate Policy Evolution
• Orthogonal Weight Normalization: Solution to Optimization over Multiple Dependent Stiefel Manifolds in Deep Neural Networks
• Recurrent Orthogonal Networks and Long-Memory Tasks
• Orthogonal Deep Neural Networks
• On Connections between Constrained Optimization and Reinforcement Learning
• Orthogonal Convolutional Neural Networks