arXiv research

Develops significance tests for neural networks without strong assumptions or excessive computation.

Paper proposes a black-box technique to generate adversarial samples.

Black-box risk scoring models permeate our lives, yet are typically proprietary or opaque. We propose Distill-and-Compare, a model distillation and comparison approach to audit such models. To gain insight into black-box models, we treat them as teachers, training transparent student models to mimic the risk scores ass…

Black-box optimization is primarily important for many compute-intensive applications, including reinforcement learning (RL), robot control, etc. This paper presents a novel theoretical framework for black-box optimization, in which our method performs stochastic update with the implicit natural gradient of an exponent…

New approach tackles non-stationary multi-agent games with black-box methods.

New method optimizes black-box functions using generative models and Wasserstein distance.

Faced with distribution shift between training and test set, we wish to detect and quantify the shift, and to correct our classifiers without test set labels. Motivated by medical diagnosis, where diseases (targets) cause symptoms (observations), we focus on label shift, where the label marginal $p(y)$ changes but the …

Deep-PrAE improves rare-event simulation for black-box systems.

Interpretable machine learning has become a strong competitor for traditional black-box models. However, the possible loss of the predictive performance for gaining interpretability is often inevitable, putting practitioners in a dilemma of choosing between high accuracy (black-box models) and interpretability (interpr…

Method optimizes diffusion model generation to meet user preferences.

In recent years, deep neural networks (DNN) have become a highly active area of research, and shown remarkable achievements on a variety of computer vision tasks. DNNs, however, are known to often make overconfident yet incorrect predictions on out-of-distribution samples, which can be a major obstacle to real-world de…

Powerful adversarial attack methods are vital for understanding how to construct robust deep neural networks (DNNs) and for thoroughly testing defense techniques. In this paper, we propose a black-box adversarial attack algorithm that can defeat both vanilla DNNs and those generated by various defense techniques develo…

Hybrid neural network infers states from black-box systems.

Analyzing large-scale, multi-experiment studies requires scientists to test each experimental outcome for statistical significance and then assess the results as a whole. We present Black Box FDR (BB-FDR), an empirical-Bayes method for analyzing multi-experiment studies when many covariates are gathered per experiment.…

Deep neural networks (DNNs) are known for their vulnerability to adversarial examples. These are examples that have undergone small, carefully crafted perturbations, and which can easily fool a DNN into making misclassifications at test time. Thus far, the field of adversarial research has mainly focused on image model…

Develops causal explanations for black-box classifiers.

Survey of algorithms for testing AI-driven CPS safety.

This paper explores how the generalization of substitute classifiers affects the success of black-box adversarial attacks.

Proposes a method to ensure accurate estimation of rare events in AI systems.

Trieste optimizes black-box functions using TensorFlow for efficient decision-making.

In science and medicine, model interpretations may be reported as discoveries of natural phenomena or used to guide patient treatments. In such high-stakes tasks, false discoveries may lead investigators astray. These applications would therefore benefit from control over the finite-sample error rate of interpretations…

Develops a method to continuously audit black-box conditional quantile forecasts.

Recently, neural networks trained as optimizers under the "learning to learn" or meta-learning framework have been shown to be effective for a broad range of optimization tasks including derivative-free black-box function optimization. Recurrent neural networks (RNNs) trained to optimize a diverse set of synthetic non-…

Spanning attack improves black-box attacks with unlabeled data.

New research shows many recent defenses against adversarial examples are ineffective against black-box attacks.

Model proposes how regulators should oversee complex algorithms in high-stakes applications.

Machine Learning systems are vulnerable to adversarial attacks and will highly likely produce incorrect outputs under these attacks. There are white-box and black-box attacks regarding to adversary's access level to the victim learning algorithm. To defend the learning systems from these attacks, existing methods in th…

Unified approach for sequence design combining likelihood-free inference and black-box optimization.

The paper tests semantic importance in opaque models using betting.

ODS improves adversarial attacks by maximizing output diversity.

NP-Attack reduces query counts for black-box adversarial attacks.

MPHD transfers knowledge across different domains for Bayesian optimization.

Black box machine learning models are currently being used for high stakes decision-making throughout society, causing problems throughout healthcare, criminal justice, and in other domains. People have hoped that creating methods for explaining these black box models will alleviate some of these problems, but trying t…

Optimizing expensive black-box systems with limited data is an extremely challenging problem. As a resolution, we present a new surrogate optimization approach by addressing two gaps in prior research -- unimportant input variables and inefficient treatment of uncertainty associated with the black-box output. We first …

New method improves black-box attacks using pre-trained models.

Understanding how a learned black box works is of crucial interest for the future of Machine Learning. In this paper, we pioneer the question of the global interpretability of learned black box models that assign numerical values to symbolic sequential data. To tackle that task, we propose a spectral algorithm for the …

Most of the work on interpretable machine learning has focused on designing either inherently interpretable models, which typically trade-off accuracy for interpretability, or post-hoc explanation systems, which lack guarantees about their explanation quality. We propose an alternative to these approaches by directly r…

Paper proposes a new black-box attack approach to minimize visual distortion.

Adapts model-based advice to stabilize black-box policies for nonlinear control.

Deep neural networks work well at approximating complicated functions when provided with data and trained by gradient descent methods. At the same time, there is a vast amount of existing functions that programmatically solve different tasks in a precise manner eliminating the need for training. In many cases, it is po…

While autonomous vehicle (AV) technology has shown substantial progress, we still lack tools for rigorous and scalable testing. Real-world testing, the $\textit{de-facto}$ evaluation method, is dangerous to the public. Moreover, due to the rare nature of failures, billions of miles of driving are needed to statisticall…

New framework for variational coresets simplifies Bayesian inference for complex models.

PyBADS optimizes complex functions quickly and reliably.

Sparse perturbations improve convergence in SZO methods for faster training.

Many deployed learned models are black boxes: given input, returns output. Internal information about the model, such as the architecture, optimisation procedure, or training data, is not disclosed explicitly as it might contain proprietary information or make the system more vulnerable. This work shows that such attri…

P-BO reduces black-box adversarial attacks by 10x with Bayesian optimization and function prior.

New approach improves black-box planning efficiency by discovering focused macros.

Efficient method for generating adversarial examples with limited query budget.