Search Results for author:
Found 27 papers, 6 papers with code
Discrete Rényi Classifiers
We prove that for a given set of marginals, the minimum Hirschfeld-Gebelein-Renyi (HGR) correlation principle introduced in [1] leads to a randomized classification rule which is shown to have a misclassification rate no larger than twice the misclassification rate of the optimal classifier.
A Minimax Approach to Supervised Learning
Given a task of predicting $Y$ from $X$, a loss function $L$, and a set of probability distributions $\Gamma$ on $(X, Y)$, what is the optimal decision rule minimizing the worst-case expected loss over $\Gamma$?
Understanding GANs: the LQG Setting
Generative Adversarial Networks (GANs) have become a popular method to learn a probability model from data.
A Spectral Approach to Generalization and Optimization in Neural Networks
The recent success of deep neural networks stems from their ability to generalize well on real data; however, Zhang et al. have observed that neural networks can easily overfit random labels.
A Convex Duality Framework for GANs
For a convex set $\mathcal{F}$, this duality framework interprets the original GAN formulation as finding the generative model with minimum JS-divergence to the distributions penalized to match the moments of the data distribution, with the moments specified by the discriminators in $\mathcal{F}$.
Generalizable Adversarial Training via Spectral Normalization
A significant portion of this gap can be attributed to the decrease in generalization performance due to adversarial training.
GANs May Have No Nash Equilibria
We discuss several numerical experiments demonstrating the existence of proximal equilibrium solutions in GAN minimax problems.
Robust Federated Learning: The Case of Affine Distribution Shifts
In such settings, the training data is often statistically heterogeneous and manifests various distribution shifts across users, which degrades the performance of the learnt model.
GAT-GMM: Generative Adversarial Training for Gaussian Mixture Models
Motivated by optimal transport theory, we design the zero-sum game in GAT-GMM using a random linear generator and a softmax-based quadratic discriminator architecture, which leads to a non-convex concave minimax optimization problem.
Train simultaneously, generalize better: Stability of gradient-based minimax learners
In this paper, we show that the optimization algorithm also plays a key role in the generalization performance of the trained minimax model.
A Wasserstein Minimax Framework for Mixed Linear Regression
We propose an optimal transport-based framework for MLR problems, Wasserstein Mixed Linear Regression (WMLR), which minimizes the Wasserstein distance between the learned and target mixture regression models.
Group-Structured Adversarial Training
Robust training methods against perturbations to the input data have received great attention in the machine learning literature.
An Optimal Transport Approach to Personalized Federated Learning
In this paper, we focus on this problem and propose a novel personalized Federated Learning scheme based on Optimal Transport (FedOT) as a learning algorithm that learns the optimal transport maps for transferring data points to a common distribution as well as the prediction model under the applied transport map.
On the Role of Generalization in Transferability of Adversarial Examples
We support our theoretical results by performing several numerical experiments showing the role of the substitute network's generalization in generating transferable adversarial examples.
On Convergence of Gradient Descent Ascent: A Tight Local Analysis
In this paper, we aim to bridge this gap by analyzing the \emph{local convergence} of general \emph{nonconvex-nonconcave} minimax problems.
Universal Adversarial Directions
Despite their great success in image recognition tasks, deep neural networks (DNNs) have been observed to be susceptible to universal adversarial perturbations (UAPs) which perturb all input samples with a single perturbation vector.
Interpretation of Neural Networks is Susceptible to Universal Adversarial Perturbations
Interpreting neural network classifiers using gradient-based saliency maps has been extensively studied in the deep learning literature.
MoreauGrad: Sparse and Robust Interpretation of Neural Networks via Moreau Envelope
Explaining the predictions of deep neural nets has been a topic of great interest in the computer vision literature.
DiffPattern: Layout Pattern Generation via Discrete Diffusion
Deep generative models dominate the existing literature in layout pattern generation.
On the Evaluation of Generative Models in Distributed Learning Tasks
First, we focus on the Fr\'echet inception distance (FID) and consider the following FID-based aggregate scores over the clients: 1) FID-avg as the mean of clients' individual FID scores, 2) FID-all as the FID distance of the trained model to the collective dataset containing all clients' data.
Provably Efficient CVaR RL in Low-rank MDPs
We study CVaR RL in low-rank MDPs with nonlinear function approximation.
An Information Theoretic Approach to Interaction-Grounded Learning
Reinforcement learning (RL) problems where the learner attempts to infer an unobserved reward from some feedback variables have been studied in several recent papers.
An Interpretable Evaluation of Entropy-based Novelty of Generative Models
The massive developments of generative model frameworks and architectures require principled methods for the evaluation of a model's novelty compared to a reference dataset or baseline generative models.
On the Inductive Biases of Demographic Parity-based Fair Learning Algorithms
Finally, we present several numerical results on the application of DP-based learning methods to standard centralized and distributed learning problems.
ChatPattern: Layout Pattern Customization via Natural Language
Existing works focus on fixed-size layout pattern generation, while the more practical free-size pattern generation receives limited attention.
Structured Gradient-based Interpretations via Norm-Regularized Adversarial Training
We show a duality relation between the regularized norms of the adversarial perturbations and gradient-based maps, based on which we design adversarial training loss functions promoting sparsity and group-sparsity properties in simple gradient maps.
Stability and Generalization in Free Adversarial Training
In this work, we study the generalization performance of adversarial training methods using the algorithmic stability framework.
