arXiv research

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

We consider the problem of controlling an unknown linear dynamical system in the presence of (nonstochastic) adversarial perturbations and adversarial convex loss functions. In contrast to classical control, the a priori determination of an optimal controller here is hindered by the latter's dependence on the yet unkno…

Paper addresses eigenvector perturbation in small eigen-gap scenarios.

Generative Intervention Models predict perturbation effects without knowing the underlying mechanisms.

The paper tackles extrapolation of gene knockouts effects on RNA counts.

New algorithm controls systems with unknown, changing losses.

A new exploration strategy for contextual bandits reduces regret and is computationally efficient.

Identifies root causes of outliers in unknown cyclic graphs.

New method identifies latent variables with sparse perturbations.

The paper shows how to learn causal representations with few environments and finite samples.

A machine learning model that generalizes well should obtain low errors on unseen test examples. Thus, if we know how to optimally perturb training examples to account for test examples, we may achieve better generalization performance. However, obtaining such perturbation is not possible in standard machine learning f…

A new method for reconstructing flows from perturbed distributions.

New method creates universal perturbations to fool neural network interpretations.

SAMS-VAE models cellular perturbations using sparse additive mechanisms.

New algorithm POO optimizes noisy, unknown-smooth functions.

Improved KSD test for better detection of differences in distributions.

Study on instability of extreme Reissner-Nordström spacetime perturbations.

Numerous empirical evidence has corroborated that the noise plays a crucial rule in effective and efficient training of neural networks. The theory behind, however, is still largely unknown. This paper studies this fundamental problem through training a simple two-layer convolutional neural network model. Although trai…

We investigate two perturbation approaches to overcome conservatism that optimism based algorithms chronically suffer from in practice. The first approach replaces optimism with a simple randomization when using confidence sets. The second one adds random perturbations to its current estimate before maximizing the expe…

New findings show learnable distributions remain learnable even with noisy or adversarial perturbations.

Machine Learning models have been shown to be vulnerable to adversarial examples, ie. the manipulation of data by a attacker to defeat a defender's classifier at test time. We present a novel probabilistic definition of adversarial examples in perfect or limited knowledge setting using prior probability distributions o…

The verification problem for neural networks is verifying whether a neural network will suffer from adversarial samples, or approximating the maximal allowed scale of adversarial perturbation that can be endured. While most prior work contributes to verifying feed-forward networks, little has been explored for verifyin…

Defense strategy improves controller robustness against adversarial attacks.

A common goal in statistics and machine learning is to learn models that can perform well against distributional shifts, such as latent heterogeneous subpopulations, unknown covariate shifts, or unmodeled temporal effects. We develop and analyze a distributionally robust stochastic optimization (DRO) framework that lea…

GnIES recovers causal structure from unknown interventions.

This paper aims to address two fundamental challenges arising in eigenvector estimation and inference for a low-rank matrix from noisy observations: (1) how to estimate an unknown eigenvector when the eigen-gap (i.e. the spacing between the associated eigenvalue and the rest of the spectrum) is particularly small; (2) …

Paper introduces metrics for evaluating multi-agent policies using best response dynamics.

This note studies a method for the efficient estimation of a finite number of unknown parameters from linear equations, which are perturbed by Gaussian noise. In case the unknown parameters have only few nonzero entries, the proposed estimator performs more efficiently than a traditional approach. The method consists o…

Wave maps with noise can lead to self-similar blowup from arbitrary initial data.

A new pricing strategy learns customer valuations without noise distribution knowledge.

TASID learns policies in high-dimensional settings with abstract simulator knowledge.

Given a measurement graph $G= (V,E)$ and an unknown signal $r \in \mathbb{R}^n$, we investigate algorithms for recovering $r$ from pairwise measurements of the form $r_i - r_j$; $\{i,j\} \in E$. This problem arises in a variety of applications, such as ranking teams in sports data and time synchronization of distribute…

We consider the problem of controlling a possibly unknown linear dynamical system with adversarial perturbations, adversarially chosen convex loss functions, and partially observed states, known as non-stochastic control. We introduce a controller parametrization based on the denoised observations, and prove that apply…

This paper solves the convergence problem for estimating MGGD parameters with a convex formulation.

New algorithm reduces prediction errors across various loss functions.

The SABR model is a stochastic volatility model not admitting a closed form solution. Hagan, Kumar, Leniewski and Woodward have obtained an approximate solution by means of perturbative techniques. A more precise approximation was found by Henry-Labordère with the heat kernel expansion method. The latter relies on deep…

We study the performance of the certainty equivalent controller on Linear Quadratic (LQ) control problems with unknown transition dynamics. We show that for both the fully and partially observed settings, the sub-optimality gap between the cost incurred by playing the certainty equivalent controller on the true system …

Paper tackles robust graph matching in dense graphs with AMP type algorithm.

There is a higher dimensional analogue of the perturbative Chern-Simons theory in the sense that a similar perturbative series as in 3-dimension, which is computed via configuration space integral, yields an invariant of higher dimensional knots (Bott-Cattaneo-Rossi invariant), which is constructed by Bott for degree 2…

New method estimates heterogeneous treatment effects with improved guarantees.

Neural networks can achieve optimal sample complexity for learning single-index models.

New algorithm learns optimal policies with just 1 episode, settling horizon-dependence in RL.

Injecting adversarial examples during training, known as adversarial training, can improve robustness against one-step attacks, but not for unknown iterative attacks. To address this challenge, we first show iteratively generated adversarial images easily transfer between networks trained with the same strategy. Inspir…

New model accounts for scale variation and noise in pairwise comparisons.

Local convolutions bias neural networks towards high-frequency adversarial examples.

This paper proposes Evolutionary Multi-objective Optimization (EMO)-based Adversarial Example (AE) design method that performs under black-box setting. Previous gradient-based methods produce AEs by changing all pixels of a target image, while previous EC-based method changes small number of pixels to produce AEs. Than…

Multi-domain learning (MDL) aims at obtaining a model with minimal average risk across multiple domains. Our empirical motivation is automated microscopy data, where cultured cells are imaged after being exposed to known and unknown chemical perturbations, and each dataset displays significant experimental bias. This p…

Face recognition system trained with noisy labels.