Adaptive multi-fidelity optimization with fast learning rates

In multi-fidelity optimization, we have access to biased approximations of varying costs of the target function. In this work, we study the setting of optimizing a locally smooth function with a limited budget Λ, where the learner has to make a trade-off between the cost and the bias of these approximations. First, we prove lower bounds for the simple regret under different assumptions on the fidelities, based on a cost-to-bias function. We then present the Kometo algorithm which achieves, with additional logarithmic factors, the same rates without any knowledge of the function smoothness and fidelity assumptions, and improves prior results. Finally, we empirically show that our algorithm outperforms prior multi-fidelity optimization methods without the knowledge of problem-dependent parameters.

[1]  Ameet Talwalkar,et al.  Hyperband: Bandit-Based Configuration Evaluation for Hyperparameter Optimization , 2016, ICLR.

[2]  Jonathan P. How,et al.  Reinforcement learning with multi-fidelity simulators , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[3]  Peter Auer,et al.  Improved Rates for the Stochastic Continuum-Armed Bandit Problem , 2007, COLT.

[4]  Liping Wang,et al.  A Strategy for Adaptive Sampling of Multi-fidelity Gaussian Process to Reduce Predictive Uncertainty , 2019, Volume 2B: 45th Design Automation Conference.

[5]  R. A. Miller,et al.  Sequential kriging optimization using multiple-fidelity evaluations , 2006 .

[6]  Rémi Munos,et al.  From Bandits to Monte-Carlo Tree Search: The Optimistic Principle Applied to Optimization and Planning , 2014, Found. Trends Mach. Learn..

[7]  Peter L. Bartlett,et al.  A simple parameter-free and adaptive approach to optimization under a minimal local smoothness assumption , 2018, ALT.

[8]  Michal Valko,et al.  General parallel optimization a without metric , 2019, ALT.

[9]  Rémi Munos,et al.  Optimistic Optimization of Deterministic Functions , 2011, NIPS 2011.

[10]  Kirthevasan Kandasamy,et al.  Noisy Blackbox Optimization with Multi-Fidelity Queries: A Tree Search Approach , 2018, AISTATS.

[11]  Rémi Munos,et al.  Stochastic Simultaneous Optimistic Optimization , 2013, ICML.

[12]  Kirthevasan Kandasamy,et al.  Multi-fidelity Gaussian Process Bandit Optimisation , 2016, J. Artif. Intell. Res..

[13]  Csaba Szepesvári,et al.  –armed Bandits , 2022 .

[14]  Kirthevasan Kandasamy,et al.  The Multi-fidelity Multi-armed Bandit , 2016, NIPS.

[15]  Yu Maruyama,et al.  Global Continuous Optimization with Error Bound and Fast Convergence , 2016, J. Artif. Intell. Res..

[16]  Kirthevasan Kandasamy,et al.  Multi-fidelity Bayesian Optimisation with Continuous Approximations , 2017, ICML.

[17]  Kirthevasan Kandasamy,et al.  Gaussian Process Bandit Optimisation with Multi-fidelity Evaluations , 2016, NIPS.

[18]  A. Hoorfar,et al.  INEQUALITIES ON THE LAMBERTW FUNCTION AND HYPERPOWER FUNCTION , 2008 .

[19]  Alexandra Carpentier,et al.  Adaptivity to Smoothness in X-armed bandits , 2018, COLT.

[20]  Yurii Nesterov,et al.  Random Gradient-Free Minimization of Convex Functions , 2015, Foundations of Computational Mathematics.

[21]  Rémi Munos,et al.  Black-box optimization of noisy functions with unknown smoothness , 2015, NIPS.

[22]  Andreas Krause,et al.  Information-Theoretic Regret Bounds for Gaussian Process Optimization in the Bandit Setting , 2009, IEEE Transactions on Information Theory.

[23]  Kirthevasan Kandasamy,et al.  Multi-Fidelity Black-Box Optimization with Hierarchical Partitions , 2018, ICML.