Discriminative Feature Feedback is a setting proposed by Dastupta et al. (2018), which provides a protocol for interactive learning based on feature explanations that are provided by a human teacher. The features distinguish between the labels of pairs of possibly similar instances. That work has shown that learning in this model can have considerable statistical and computational advantages over learning in standard label-based interactive learning models.
In this work, we provide new robust interactive learning algorithms for the Discriminative Feature Feedback model, with mistake bounds that are significantly lower than those of previous robust algorithms for this setting. In the adversarial setting, we reduce the dependence on the number of protocol exceptions from quadratic to linear. In addition, we provide an algorithm for a slightly more restricted model, which obtains an even smaller mistake bound for large models with many exceptions.
In the stochastic setting, we provide the first algorithm that converges to the exception rate with a polynomial sample complexity. Our algorithm and analysis for the stochastic setting involve a new construction that we call Feature Influence, which may be of wider applicability.
Siamese Neural Network with Joint Bayesian Model Structure for Speaker Verification
National Institute of Information and Communications Technology(国立信息与通信技术研究所);Research Center for Information Technology Innovation, Academic Sinica(中央研究院资讯科技创新研究中心)
Improved High-Probability Regret for Adversarial Bandits with Time-Varying Feedback Graphs
University of Southern California(南加州大学);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)
我们研究了在$T$回合上具有时变反馈图的对抗性$K$武装匪徒的高概率后悔界。对于一般的强可观测图,我们提出了一个算法,该算法以高概率获得最优遗憾$\widetilde{\mathcal{O}}((\sum_{t=1}^Tα_t)^{1/2}+\max_{t\in[T]}α_t)$,其中$α_t$是反馈图在循环$t$的独立数.与现有的最佳结果[Neu,2015]相比,它只考虑所有节点都具有自循环的图,我们的结果不仅更普遍,而且重要的是还消除了任何$\text{poly}(K)$依赖性,这些依赖性对于上下文强盗等应用程序来说可能非常大。此外,我们还开发了第一个算法,该算法实现了弱可观测图的最佳高概率遗憾界,甚至改进了[Alon et al.,2015]通过使用精细分析删除$\mathcal{O}(\sqrt{KT})$项。我们的算法是基于在线镜像下降框架,但重要的是与几种技术的创新组合。值得注意的是,虽然早期的作品使用乐观偏置损失估计实现高概率界,我们发现重要的是使用一个悲观的节点没有自我循环在一个强可观察的图。
Discriminative feature generation for classification of imbalanced data
Korea Institute of Science and Technology Europe Forschungsgesellschaft mbH(韩国科学技术院欧洲研究所);TU Kaiserslautern(凯泽斯劳滕工业大学);German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心)
Improved Regret for Efficient Online Reinforcement Learning with Linear Function Approximation
Tel Aviv University(特拉维夫大学);Google Research(谷歌研究院)
我们研究了具有线性函数近似和逆向变化代价函数的强化学习,这种设置大多被认为是在完全信息反馈或探索性条件等简化假设下进行的。我们提出了一种计算效率高的策略优化算法,用于具有挑战性的未知动态和强盗反馈的一般设置,在计算探索奖金的辅助MDP中结合了镜像下降和最小二乘策略评估。我们的算法获得了$\widetilde O(K^{6/7})$遗憾界,在这种情况下明显优于现有技术的$\widetilde O (K^{14/15})$。此外,我们提出了一个版本的相同的算法的假设下,环境的模拟器可供学习者(但否则没有探索性的假设),并证明它获得了国家的最先进的遗憾$\widetilde O (K^{2/3})$。
University of Science, Ho Chi Minh City(胡志明市科学大学);Vietnam National University, Ho Chi Minh City(胡志明市越南国家大学);University of Liverpool(利物浦大学);University of Information Technology(信息技术大学)