arXiv research

Reward-poisoning attacks can force RL agents to learn bad policies, and we categorize and quantify their feasibility.

Deep neural networks are vulnerable to adversarial examples, i.e., carefully-perturbed inputs aimed to mislead classification. This work proposes a detection method based on combining non-linear dimensionality reduction and density estimation techniques. Our empirical findings show that the proposed approach is able to…

New defense method inspired by encryption improves visual classification accuracy.

Many current approaches to the design of intrusion detection systems apply feature selection in a static, non-adaptive fashion. These methods often neglect the dynamic nature of network data which requires to use adaptive feature selection techniques. In this paper, we present a simple technique based on incremental le…

Paper tackles adaptive deletion of data points from trained models.

New study shows non-adaptive trials can be outperformed by adaptive designs in treatment selection.

Sublinear algorithms detect cliques in graphs with high probability.

Study learns random hypergraphs with queries, improving on previous results.

New protocols show 1-bit mean estimation can be order-optimal without interaction.

Study shows rates for Laplacian-eigenmap methods in nonparametric regression.

In practice, the data distribution at test time often differs, to a smaller or larger extent, from that of the original training data. Consequentially, the so-called source classifier, trained on the available labelled data, deteriorates on the test, or target, data. Domain adaptive classifiers aim to combat this probl…

New algorithm reduces interventional strategy complexity for causal graph discovery.

New method infers viral load from pooled tests.

New algorithm solves stochastic optimization problems with unknown gradients.

Optimizes group testing for COVID-19 to reduce test numbers.

New algorithms learn MNL weights efficiently for any slate size.

Adaptive sampling results in dramatic improvements in the recovery of sparse signals in white Gaussian noise. A sequential adaptive sampling-and-refinement procedure called Distilled Sensing (DS) is proposed and analyzed. DS is a form of multi-stage experimental design and testing. Because of the adaptive nature of the…

Adversarial examples are a pervasive phenomenon of machine learning models where seemingly imperceptible perturbations to the input lead to misclassifications for otherwise statistically accurate models. In this paper we study how the choice of optimization algorithm influences the robustness of the resulting classifie…

Adaptive data analysis is frequently criticized for its pessimistic generalization guarantees. The source of these pessimistic bounds is a model that permits arbitrary, possibly adversarial analysts that optimally use information to bias results. While being a central issue in the field, still lacking are notions of na…

Recent breakthrough results in compressive sensing (CS) have established that many high dimensional signals can be accurately recovered from a relatively small number of non-adaptive linear observations, provided that the signals possess a sparse representation in some basis. Subsequent efforts have shown that the perf…

We show that for the problem of testing if a matrix $A \in F^{n \times n}$ has rank at most $d$, or requires changing an $ε$-fraction of entries to have rank at most $d$, there is a non-adaptive query algorithm making $\widetilde{O}(d^2/ε)$ queries. Our algorithm works for any field $F$. This improves upon the previous…

Adaptive networks improve model robustness through conditional normalization.

In domain adaptation, classifiers with information from a source domain adapt to generalize to a target domain. However, an adaptive classifier can perform worse than a non-adaptive classifier due to invalid assumptions, increased sensitivity to estimation errors or model misspecification. Our goal is to develop a doma…

Algorithm estimates principal eigenvector with adaptive sensing, improving over non-adaptive methods.

We solve non-Markovian optimal switching problems in discrete time on an infinite horizon, when the decision maker is risk aware and the filtration is general, and establish existence and uniqueness of solutions for the associated reflected backward stochastic difference equations. An example application to hydropower …

New findings on complexity limits in fixed budget bandit identification.

We study the group testing problem with non-adaptive randomized algorithms. Several models have been discussed in the literature to determine how to randomly choose the tests. For a model ${\cal M}$, let $m_{\cal M}(n,d)$ be the minimum number of tests required to detect at most $d$ defectives within $n$ items, with su…

We study the problem of finding the most mutually correlated arms among many arms. We show that adaptive arms sampling strategies can have significant advantages over the non-adaptive uniform sampling strategy. Our proposed algorithms rely on a novel correlation estimator. The use of this accurate estimator allows us t…

Optimization lies at the heart of machine learning and signal processing. Contemporary approaches based on the stochastic gradient method are non-adaptive in the sense that their implementation employs prescribed parameter values that need to be tuned for each application. This article summarizes recent research and mo…

AdaGrad-Norm achieves optimal convergence rates for non-convex objectives without tuning.

Sampling from distributions to find the one with the largest mean arises in a broad range of applications, and it can be mathematically modeled as a multi-armed bandit problem in which each distribution is associated with an arm. This paper studies the sample complexity of identifying the best arm (largest mean) in a m…

New algorithms improve online prediction from experts with privacy constraints.

This paper studies the sample complexity of searching over multiple populations. We consider a large number of populations, each corresponding to either distribution P0 or P1. The goal of the search problem studied here is to find one population corresponding to distribution P1 with as few samples as possible. The main…

"Feint Attack", as a new type of APT attack, has become the focus of attention. It adopts a multi-stage attacks mode which can be concluded as a combination of virtual attacks and real attacks. Under the cover of virtual attacks, real attacks can achieve the real purpose of the attacker, as a result, it often caused hu…

This paper studies adversarial attacks on Gaussian process bandits.

This paper proposes a simple adaptive sensing and group testing algorithm for sparse signal recovery. The algorithm, termed Compressive Adaptive Sense and Search (CASS), is shown to be near-optimal in that it succeeds at the lowest possible signal-to-noise-ratio (SNR) levels, improving on previous work in adaptive comp…

Subpopulation attacks poison data to misclassify naturally distributed points.

We consider the problem of group testing with sum observations and noiseless answers, in which we aim to locate multiple objects by querying the number of objects in each of a sequence of chosen sets. We study a probabilistic setting with entropy loss, in which we assume a joint Bayesian prior density on the locations …

Spanning attack improves black-box attacks with unlabeled data.

Headless attacks bypass classification heads to fool transfer learning models.

This paper explores evasion attacks against Bayesian models.

Adversarial attacks pose a threat to deep neural networks, especially in safety-critical applications.

RayS attack improves hard-label adversarial attacks by reducing query complexity and identifying false robust models.

Depending on how much information an adversary can access to, adversarial attacks can be classified as white-box attack and black-box attack. For white-box attack, optimization-based attack algorithms such as projected gradient descent (PGD) can achieve relatively high attack success rates within moderate iterates. How…

Due to the broad range of applications of stochastic multi-armed bandit model, understanding the effects of adversarial attacks and designing bandit algorithms robust to attacks are essential for the safe applications of this model. In this paper, we introduce a new class of attack named action-manipulation attack. In …

Control policies, trained using the Deep Reinforcement Learning, have been recently shown to be vulnerable to adversarial attacks introducing even very small perturbations to the policy input. The attacks proposed so far have been designed using heuristics, and build on existing adversarial example crafting techniques …

There has been an ongoing cycle where stronger defenses against adversarial attacks are subsequently broken by a more advanced defense-aware attack. We present a new approach towards ending this cycle where we "deflect'' adversarial attacks by causing the attacker to produce an input that semantically resembles the att…

New attacks reveal membership in label-only ML models.