Batch Reinforcement Learning Batch Reinforcement Learning Coach supports Batch Reinforcement Learning, where learning is based solely on a (fixed) batch of data.In Batch RL, we...

强化学习 (DQN) 教程 回放内存 DQN 算法 Q-网络 获取输入 训练 超参数和配置 训练循环 强化学习 (DQN) 教程 译者：平淡的天 作者 : Adam Paszke 本教程将展示如何使用 PyTorch 在OpenAI Gym 的任务集上训练一个深度Q学习 (DQN) 智能点。 任务 智能点需要决定...

Benchmarks Benchmarks Reinforcement learning is a developing field, and so far it has been particularly difficult to reproduce some of theresults published in the original pape...

术语中英对照表 术语中英对照表 张量 Tensor 变量 Variable 操作 Operation 优化器 Optimizer 层 Layer 损失函数 Loss Function 评估指标 Metrics 多层感知机 Multilayer Perceptron, MLP 卷积神经网络 Convolutional ...

Environments Environments Coach supports a large number of environments which can be solved using reinforcement learning.To find a detailed documentation of the environments AP...

Auto-scaling with Broodlord mode A ‘simple’ example –vassal-sos Manually asking for reinforcement Under the hood (or: hacking broodlord mode) Auto-scaling with Broodlord mo...

Features Features Features Algorithms Environments Benchmarks Batch Reinforcement Learning

Algorithms Algorithms Coach supports many state-of-the-art reinforcement learning algorithms, which are separated into three main classes -value optimization, policy optimizati...

Selecting an Algorithm 1. Does your environment have a discrete or continuous action space? 2. Is collecting more samples from your environment painful? 3. Do you have a simulato...

Dopamine What’s new Instructions Install via source Ubuntu Mac OS X Running tests Install as a library Running tests References Giving credit Dopamine Dopamine i...