Structured Evolution with Compact Architectures for Scalable Policy Optimization
원문 초록
We present a new method of blackbox optimization via gradient approximation with the use of structured random orthogonal matrices, providing more accurate estimators than baselines and with provable theoretical guarantees. We show that this algorithm can be successfully applied to learn better quality compact policies than those using standard gradient estimation techniques. The compact policies we learn have several advantages over unstructured ones, including faster training algorithms and faster inference. These benefits are important when the policy is deployed on real hardware with limited resources. Further, compact policies provide more scalable architectures for derivative-free optimization (DFO) in high-dimensional spaces. We show that most robotics tasks from the OpenAI Gym can be solved using neural networks with less than 300 parameters, with almost linear time complexity of the inference phase, with up to 13x fewer parameters relative to the Evolution Strategies (ES) algorithm introduced by Salimans et al. (2017). We do not need heuristics such as fitness shaping to learn good quality policies, resulting in a simple and theoretically motivated training mechanism.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
CreatiPoster: Towards Editable and Controllable Multi-Layer Graphic Design Generation
CharaConsist: Fine-Grained Consistent Character Generation
NeoBabel: A Multilingual Open Tower for Visual Generation
Ovis-U1 Technical Report
Rethink Sparse Signals for Pose-guided Text-to-image Generation
XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulation
토론
이 논문과 연결된 포럼 글이 아직 없습니다.