A provable control of sensitivity of neural networks through a direct parameterization of the overall bi-Lipschitzness
While neural networks can enjoy an outstanding flexibility and exhibit unprecedented performance, the mechanism behind their behavior is still not well-understood. To tackle this fundamental challenge, researchers have tried to restrict and manipulate some of their properties in order to gain new insights and better control on them. Especially, throughout the past few years, the concept of \emph{bi-Lipschitzness} has been proved as a beneficial inductive bias in many areas. However, due to its complexity, the design and control of bi-Lipschitz architectures are falling behind, and a model that is precisely designed for bi-Lipschitzness realizing a direct and simple control of the constants along with solid theoretical analysis is lacking. In this work, we investigate and propose a novel framework for bi-Lipschitzness that can achieve such a clear and tight control based on convex neural networks and the Legendre-Fenchel duality. Its desirable properties are illustrated with concrete experiments. We also apply this framework to uncertainty estimation and monotone problem settings to illustrate its broad range of applications.
Code (1)
Tasks
Inductive BiasSensitivitySimilar Papers 제목 키워드 기반
On the sensitivity of pose estimation neural networks: rotation parameterizations, Lipschitz constants, and provable bounds
In this paper, we approach the task of determining sensitivity bounds for pose estimation neural networks. This task is particularly challenging as it requires characterizing the sensitivity of 3D rotations. We develop a…
Pose EstimationSensitivityLearning Control Barrier Functions from Expert Demonstrations
Inspired by the success of imitation and inverse reinforcement learning in replicating expert behavior through optimal control, we propose a learning based approach to safe controller synthesis based on control barrier f…
Reinforcement LearningHalf-Gain Tuning for Active Disturbance Rejection Control
A new tuning rule is introduced for linear active disturbance rejection control (ADRC), which results in similar closed-loop dynamics as the commonly employed bandwidth parameterization design, but with lower feedback ga…
SensitivityFrequency Domain Auto-tuning of Structured LPV Controllers for High-Precision Motion Control
Motion systems are a vital part of many industrial processes. However, meeting the increasingly stringent demands of these systems, especially concerning precision and throughput, requires novel control design methods th…
PAC-Bayesian Certificates for Quadratic Closed-Loop Control
PAC-Bayesian bounds provide finite-sample guarantees for data-dependent randomized predictors, but applying them to learning-based control is difficult because the natural objective is a quadratic trajectory cost. Such l…