arXiv research

Adversarial training achieves optimal test error for shallow networks.

New minimax optimal learner for robust predictors against adversarial examples.

New method learns optimal prediction strategies in adversarial games.

While adversarial training can improve robust accuracy (against an adversary), it sometimes hurts standard accuracy (when there is no adversary). Previous work has studied this tradeoff between standard and robust accuracy, but only in the setting where no predictor performs well on both objectives in the infinite data…

This work precisely characterizes and improves the tradeoff between robustness and accuracy in linear regression.

Study robust learning without knowing perturbation sets, using interactions with attackers.

New trade-off found between accuracy and adversarial robustness in regression.

New approach improves robustness of deep neural networks without overfitting.

New framework improves adversarial robustness in one-stage L2D.

Paper develops robust Bayesian models for linear regression under adversarial perturbations.

This paper proposes a new algorithmic framework, predictor-verifier training, to train neural networks that are verifiable, i.e., networks that provably satisfy some desired input-output properties. The key idea is to simultaneously train two networks: a predictor network that performs the task at hand,e.g., predicting…

Study shows how to improve contextual bandits with loss predictors.

Simple algorithm achieves distance to calibration error of at most 2√T+1.

New method minimizes regret in AMDP with high probability.

This work addresses local fairness in machine learning models.

Many predicted structured objects (e.g., sequences, matchings, trees) are evaluated using the F-score, alignment error rate (AER), or other multivariate performance measures. Since inductively optimizing these measures using training data is typically computationally difficult, empirical risk minimization of surrogate …

This thesis explores robust machine learning against adversarial examples.

Adversarial training is a useful approach to promote the learning of transferable representations across the source and target domains, which has been widely applied for domain adaptation (DA) tasks based on deep neural networks. Until very recently, existing adversarial domain adaptation (ADA) methods ignore the usefu…

We present a predictor-corrector framework, called PicCoLO, that can transform a first-order model-free reinforcement or imitation learning algorithm into a new hybrid method that leverages predictive models to accelerate policy learning. The new "PicCoLOed" algorithm optimizes a policy by recursively repeating two ste…

We study the question of learning an adversarially robust predictor. We show that any hypothesis class $\mathcal{H}$ with finite VC dimension is robustly PAC learnable with an improper learning rule. The requirement of being improper is necessary as we exhibit examples of hypothesis classes $\mathcal{H}$ with finite VC…

Generative Adversarial Networks (GAN) (Goodfellow et al., 2014) are an effective method for training generative models of complex data such as natural images. However, they are notoriously hard to train and can suffer from the problem of missing modes where the model is not able to produce examples in certain regions o…

Boosts barely robust learners to be more adversarially robust.

Study adversarial perturbations in classification, analyzing learning and certification.

Understanding optimal prompts for binary sequence predictors is challenging.

Two approaches scale up DNN optimization for diverse edge devices.

Post-processing predictors reduces calibration errors for decision-making.

Proposes BATer for improved adversarial example detection.

Predict covariance from features using convex optimization.

Optimal trading strategy with predictor and costs, derived equations and shape.

We resolve the open problem of optimal sample complexity for multicalibration and deterministic predictors.

We study online linear regression problems in a distributed setting, where the data is spread over a network. In each round, each network node proposes a linear predictor, with the objective of fitting the \emph{network-wide} data. It then updates its predictor for the next round according to the received local feedbac…

MALT improves adversarial attacks by targeting classes more efficiently.

Optimizing proper loss yields calibrated models under specific conditions.

The paper proposes multicalibration to improve matching in graphs with imperfect predictors.

The paper develops predictors for functional data on manifolds.

We introduce a framework for dynamic adversarial discovery of information (DADI), motivated by a scenario where information (a feature set) is used by third parties with unknown objectives. We train a reinforcement learning agent to sequentially acquire a subset of the information while balancing accuracy and fairness …

The standard risk minimization paradigm of machine learning is brittle when operating in environments whose test distributions are different from the training distribution due to spurious correlations. Training on data from many environments and finding invariant predictors reduces the effect of spurious features by co…

New loss function reduces outage probability in ML-assisted resource allocation.

Forecasting a time series from multivariate predictors constitutes a challenging problem, especially using model-free approaches. Most techniques, such as nearest-neighbor prediction, quickly suffer from the curse of dimensionality and overfitting for more than a few predictors which has limited their application mostl…

APQ jointly optimizes neural architecture, pruning, and quantization for efficient inference.

Methods of transfer learning try to combine knowledge from several related tasks (or domains) to improve performance on a test task. Inspired by causal methodology, we relax the usual covariate shift assumption and assume that it holds true for a subset of predictor variables: the conditional distribution of the target…

NPENAS improves neural architecture search efficiency and accuracy.

Study shows how pretraining robustness transfers to downstream tasks.

We identify a trade-off between robustness and accuracy that serves as a guiding principle in the design of defenses against adversarial examples. Although this problem has been widely studied empirically, much remains unknown concerning the theory underlying this trade-off. In this work, we decompose the prediction er…

A graph VAE framework optimizes neural architectures in a continuous space.

Study evaluates 31 performance predictors in NAS, recommending best for different settings.

A faster method for optimizing DNA and protein sequences using machine learning.

Ensemble methods that average over a collection of independent predictors that are each limited to a subsampling of both the examples and features of the training data command a significant presence in machine learning, such as the ever-popular random forest, yet the nature of the subsampling effect, particularly of th…