arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2301.01348cs.LGcs.AIcs.RO
收藏

DADAgger:分歧增强的数据集聚合

DADAgger: Disagreement-Augmented Dataset Aggregation

Akash Haridas, Karim Hamadeh, Samarendra Chandan Bindu Dash

摘要:

Dagger是一种模仿算法,它通过查询专家在训练过程中遇到的所有样本来聚合其原始数据集。为了减少查询的样本数量,我们提出了一个修改DAGGER,称为DADAgger,它只查询专家的状态-动作对是分布(OOD)。OOD状态通过测量每个状态上的模型集合的动作预测的方差来识别,我们使用dropout来模拟。在Car Racing和Half Cheetah环境中的测试实现了与Dagger相当的性能,但减少了专家查询,并且比随机抽样基线性能更好。我们还表明,我们的算法可以用来构建高效,平衡的训练数据集,运行没有初始数据,只查询专家来解决不确定性。

英文摘要:

DAgger is an imitation algorithm that aggregates its original datasets by querying the expert on all samples encountered during training. In order to reduce the number of samples queried, we propose a modification to DAgger, known as DADAgger, which only queries the expert for state-action pairs that are out of distribution (OOD). OOD states are identified by measuring the variance of the action predictions of an ensemble of models on each state, which we simulate using dropout. Testing on the Car Racing and Half Cheetah environments achieves comparable performance to DAgger but with reduced expert queries, and better performance than a random sampling baseline. We also show that our algorithm may be used to build efficient, well-balanced training datasets by running with no initial data and only querying the expert to resolve uncertainty.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑