Architecture Selection via the Trade-off Between Accuracy and Robustness

📅 2019-06-04
🏛️ arXiv.org
📈 Citations: 3
Influential: 0
📄 PDF

career value

194K/year
🤖 AI Summary
This work addresses the fundamental trade-off between standard accuracy and adversarial robustness in supervised learning. Methodologically, it introduces the first architecture-level accuracy–robustness trade-off curve, quantifying the inverse relationship between these objectives across diverse neural network architectures; defines a sensitivity influence function to theoretically characterize the stability of optimal solutions under adversarial perturbations; and reveals—via theoretical analysis of overparameterized linear models—that adversarial training implicitly regularizes model dynamics, interpolating between L₁ (LASSO) and L₂ (ridge regression) behaviors. The approach integrates rigorous theoretical analysis, influence-function-based modeling, and extensive empirical evaluation across fully connected, deep, and varying-width networks. Results consistently validate the existence and structure of the trade-off, providing an interpretable, predictive theoretical foundation for principled neural architecture selection.
📝 Abstract
We provide a general framework for characterizing the trade-off between accuracy and robustness in supervised learning. We propose a method and define quantities to characterize the trade-off between accuracy and robustness for a given architecture, and provide theoretical insight into the trade-off. Specifically we introduce a simple trade-off curve, define and study an influence function that captures the sensitivity, under adversarial attack, of the optima of a given loss function. We further show how adversarial training regularizes the parameters in an over-parameterized linear model, recovering the LASSO and ridge regression as special cases, which also allows us to theoretically analyze the behavior of the trade-off curve. In experiments, we demonstrate the corresponding trade-off curves of neural networks and how they vary with respect to factors such as number of layers, neurons, and across different network structures. Such information provides a useful guideline to architecture selection.
Problem

Research questions and friction points this paper is trying to address.

Characterizing accuracy-robustness trade-off in supervised learning
Analyzing adversarial attack sensitivity via influence functions
Studying architecture selection guidelines for neural networks
Innovation

Methods, ideas, or system contributions that make the work stand out.

Framework for accuracy-robustness trade-off analysis
Influence function measures adversarial sensitivity
Adversarial training links to LASSO, ridge regression
🔎 Similar Papers
No similar papers found.