图 书 馆
广义无悔问题的机会策略Opportunistic Strategies for Generalized No-Regret Problems |
|
| 课程网址: | http://videolectures.net/colt2013_bernstein_problems/ |
| 主讲教师: | Andrey Bernstein |
| 开课单位: | 以色列理工学院 |
| 开课时间: | 2013-08-09 |
| 课程语种: | 英语 |
| 中文简介: |  |
| 课程简介: | No-regret algorithms has played a key role in on-line learning and prediction problems. In this paper, we focus on a generalized no-regret problem with vector-valued rewards, defined in terms of a desired reward set of the agent. For each mixed action q of the opponent, the agent has a set R(q) where the average reward should reside. In addition, the agent has a response mixed action p which brings the expected reward under these two actions, r(p,q), to R(q). If a strategy of the agent ensures that the average reward converges to R(q¯n), where q¯n is the empirical distribution of the opponent’s actions, for any strategy of the opponent, we say that it is a no-regret strategy with respect to R(q). The standard no-regret problem is obtained as a special case for scalar rewards and R(q) = r∈R:r≥r(q), where r(q)=maxpr(p,q). For this problem, the multifunction R(q) is convex, and no-regret strategies can be devised. Our main interest in this paper is in cases where this convexity property does not hold. The best that can be guaranteed in general then is the convergence of the average reward to Rc(q¯n), the convex hull of R(q¯n). However, as the game unfolds, it may turn out that the opponent’s choices of actions are limited in some way. If these restrictions were known in advance, the agent could possibly ensure convergence of the average reward to some desired subset of Rc(q¯n), or even approach R(q¯n) itself. We formulate appropriate goals for opportunistic no-regret strategies, in the sense that they may exploit such limitations on the opponent’s action sequence in an on-line manner, without knowing them beforehand. As the main technical tool, we propose a class of approachability algorithms that rely on a calibrated forecast of the opponent’s actions, which are opportunistic in the above mentioned sense. As an application, we consider the online no-regret problem with average cost constraints, introduced in Mannor, Tsitsiklis, and Yu (2009), where the best-response-in-hindsight is not generally attainable, but only its convex relaxation. Our proposed algorithm applied to this problem does attain the best-response-in-hindsight if the opponent’s play happens to be stationary (either in terms of its mixed actions, or the empirical frequencies of its pure actions). |
| 关 键 词: | 无悔算法; 在线学习; 预测问题; 向量价值; 期望奖励集; 标量问题; 可接近性算法 |
| 课程来源: | 视频讲座网公开课 |
| 最后编审: | 2019-05-26：cwx |
| 阅读次数: | 86 |