Publications de Tobias Sutter | EPFL Graph Search

A Pareto Dominance Principle for Data-Driven Optimization

We propose a statistically optimal approach to construct data-driven decisions for stochastic optimization problems. Fundamentally, a data-driven decision is simply a function that maps the available training data to a feasible action. It can always be exp ...

2024

End-to-End Learning for Stochastic Optimization: A Bayesian Perspective

Daniel Kuhn, Tobias Sutter, Yves Rychener

We develop a principled approach to end-to-end learning in stochastic optimization. First, we show that the standard end-to-end learning algorithm admits a Bayesian interpretation and trains a posterior Bayes action map. Building on the insights of this an ...

2023

Efficient Learning of a Linear Dynamical System with Stability Guarantees

Daniel Kuhn, Tobias Sutter, Wouter Jongeneel

We propose a principled method for projecting an arbitrary square matrix to the non- convex set of asymptotically stable matrices. Leveraging ideas from large deviations theory, we show that this projection is optimal in an information-theoretic sense and ...

2023

Topological Linear System Identification via Moderate Deviations Theory

Daniel Kuhn, Tobias Sutter, Wouter Jongeneel

Two dynamical systems are topologically equivalent when their phase-portraits can be morphed into each other by a homeomorphic coordinate transformation on the state space. The induced equivalence classes capture qualitative properties such as stability or ...

2022

Distributionally Robust Optimization with Markovian Data

Daniel Kuhn, Tobias Sutter, Mengmeng Li

We study a stochastic program where the probability distribution of the uncertain problem parameters is unknown and only indirectly observed via finitely many correlated samples generated by an unknown Markov chain with d states. We propose a data-driven d ...

JMLR-JOURNAL MACHINE LEARNING RESEARCH2021

Robust Generalization despite Distribution Shift via Minimum Discriminating Information

Daniel Kuhn, Tobias Sutter, Andreas Krause

Training models that perform well under distribution shifts is a central challenge in machine learning. In this paper, we introduce a modeling framework where, in addition to training data, we have partial structural knowledge of the shifted test distribut ...

2021

Generalized maximum entropy estimation

Peyman Mohajerin Esfahani, John Lygeros, Tobias Sutter

We consider the problem of estimating a probability distribution that maximizes the entropy while satisfying a finite number of moment constraints, possibly corrupted by noise. Based on duality of convex programming, we present a novel approximation scheme ...

2019

From Infinite to Finite Programs: Explicit Error Bounds with Applications to Approximate Dynamic Programming

Daniel Kuhn, Peyman Mohajerin Esfahani, John Lygeros, Tobias Sutter

We consider linear programming (LP) problems in infinite dimensional spaces that are in general computationally intractable. Under suitable assumptions, we develop an approximation bridge from the infinite-dimensional LP to tractable finite convex programs ...

2018