Online Least Squares Estimation with Self-Normalized Processes: An Application to Bandit Problems

The analysis of online least squares estimation is at the heart of many stochastic sequential decision-making problems. We employ tools from the self-normalized processes to provide a simple and self-contained proof of a tail bound of a vector-valued martingale. We use the bound to construct new tighter confidence sets for the least squares estimate. We apply the confidence sets to several online decision problems, such as the multi-armed and the linearly parametrized bandit problems. The confidence sets are potentially applicable to other problems such as sleeping bandits, generalized linear bandits, and other linear control problems. We improve the regret bound of the Upper Confidence Bound (UCB) algorithm of Auer et al. (2002) and show that its regret is with high-probability a problem dependent constant. In the case of linear bandits (Dani et al., 2008), we improve the problem dependent bound in the dimension and number of time steps. Furthermore, as opposed to the previous result, we prove that our bound holds for small sample sizes, and at the same time the worst case bound is improved by a logarithmic factor and the constant is improved.

[1]  H. Robbins,et al.  Boundary Crossing Probabilities for the Wiener Process and Sample Sums , 1970 .

[2]  D. Freedman.  On Tail Probabilities for Martingales , 1975 .

[3]  H. Robbins,et al.  Strong consistency of least squares estimates in multiple regression. , 1979, Proceedings of the National Academy of Sciences of the United States of America.

[4]  T. Lai,et al.  Least Squares Estimates in Stochastic Regression Models with Applications to Identification and Control of Dynamic Systems , 1982 .

[5]  V. N. Bogaevski,et al.  Matrix Perturbation Theory , 1991 .

[6]  T. Lai,et al.  Self-Normalized Processes: Limit Theory and Statistical Applications , 2001 .

[7]  Peter Auer,et al.  Using Confidence Bounds for Exploitation-Exploration Trade-offs , 2003, J. Mach. Learn. Res..

[8]  Peter Auer,et al.  Finite-time Analysis of the Multiarmed Bandit Problem , 2002, Machine Learning.

[9]  T. Lai,et al.  SELF-NORMALIZED PROCESSES: EXPONENTIAL INEQUALITIES, MOMENT BOUNDS AND ITERATED LOGARITHM LAWS , 2004, math/0410102.

[10]  Gábor Lugosi,et al.  Prediction, learning, and games , 2006 .

[11]  H. Robbins.  Some aspects of the sequential design of experiments , 1952 .

[12]  Thomas P. Hayes,et al.  Stochastic Linear Optimization under Bandit Feedback , 2008, COLT.

[13]  Aurélien Garivier,et al.  On Upper-Confidence Bound Policies for Non-Stationary Bandit Problems , 2008 .

[14]  Csaba Szepesvári,et al.  Online Optimization in X-Armed Bandits , 2008, NIPS.

[15]  John N. Tsitsiklis,et al.  Linearly Parameterized Bandits , 2008, Math. Oper. Res..

[16]  Varun Grover,et al.  Active learning in heteroscedastic noise , 2010, Theor. Comput. Sci..

[17]  T. L. Lai Andherbertrobbins.  Asymptotically Efficient Adaptive Allocation Rules , 2022 .