Deep reinforcement learning with domain randomization learns a control policy in various simulations with randomized physical and sensor model parameters to become transferable to the real world in a zero-shot setting. However, a huge number of samples are often required to learn an effective policy when the range of randomized parameters is extensive due to the instability of policy updates. To alleviate this problem, we propose a sample-efficient method named cyclic policy distillation (CPD). CPD divides the range of randomized parameters into several small sub-domains and assigns a local policy to each one. Then local policies are learned while cyclically transitioning to sub-domains. CPD accelerates learning through knowledge transfer based on expected performance improvements. Finally, all of the learned local policies are distilled into a global policy for sim-to-real transfers. CPD's effectiveness and sample efficiency are demonstrated through simulations with four tasks (Pendulum from OpenAIGym and Pusher, Swimmer, and HalfCheetah from Mujoco), and a real-robot, ball-dispersal task. We published code and videos from our experiments at https://github.com/yuki-kadokawa/cyclic-policy-distillation.
翻译:使用域隨機化的深度強化學習可通過對物理和傳感器模型參數進行隨機優化來學習控制策略以在零增益設置下轉移到現實世界。 然而,由於策略更新的不穩定性,當隨機化參數的範圍很大時,通常需要大量樣本才能學習有效的策略。 為了減輕此問題,我們提出了一種稱為“輪廓策略提煉”的樣本有效方法(CPD)。 CPD將隨機化參數的範圍劃分為幾個小的子域,並為每個子域分配一個本地策略。 然後,在循環轉換到子域的情況下學習本地策略。 CPD通過基於預期的性能提高的知識轉移加速學習。 最後,所有學習到的本地策略都被提煉為全局策略以進行模擬到現實的轉移。 通過Pendulum from OpenAIGym和Pusher,Swimmer和HalfCheetah from Mujoco等四個任務的模擬以及一個真實機器人的球分散任務,我們證明了CPD的有效性和樣本效率。 我們在https://github.com/yuki-kadokawa/cyclic-policy-distillation上發布了代碼和實驗視頻。