Zobrazeno 1 - 10
of 35 564
pro vyhledávání: '"Vardi"'
Recent Vision-Language Models (VLMs) have demonstrated remarkable capabilities in visual understanding and reasoning, and in particular on multiple-choice Visual Question Answering (VQA). Still, these models can make distinctly unnatural errors, for
Externí odkaz:
http://arxiv.org/abs/2501.01371
Validating the behavior of autonomous Cyber-Physical Systems (CPS) and Artificial Intelligence (AI) agents, which rely on automated controllers, is an objective of great importance. In recent years, Neural-Network (NN) controllers have been demonstra
Externí odkaz:
http://arxiv.org/abs/2412.17992
We study the problem of realizing strategies for an LTLf goal specification while ensuring that at least an LTLf backup specification is satisfied in case of unreliability of certain input variables. We formally define the problem and characterize it
Externí odkaz:
http://arxiv.org/abs/2412.14728
Screening questionnaires are used in medicine as a diagnostic aid. Creating them is a long and expensive process, which could potentially be improved through analysis of social media posts related to symptoms and behaviors prior to diagnosis. Here we
Externí odkaz:
http://arxiv.org/abs/2411.11048
We introduce LTLf+ and PPLTL+, two logics to express properties of infinite traces, that are based on the linear-time temporal logics LTLf and PPLTL on finite traces. LTLf+/PPLTL+ use levels of Manna and Pnueli's LTL safety-progress hierarchy, and th
Externí odkaz:
http://arxiv.org/abs/2411.09366
We study the implicit bias of the general family of steepest descent algorithms, which includes gradient descent, sign descent and coordinate descent, in deep homogeneous neural networks. We prove that an algorithm-dependent geometric margin starts i
Externí odkaz:
http://arxiv.org/abs/2410.22069
We study the overfitting behavior of fully connected deep Neural Networks (NNs) with binary weights fitted to perfectly classify a noisy training set. We consider interpolation using both the smallest NN (having the minimal number of weights) and a r
Externí odkaz:
http://arxiv.org/abs/2410.19092
The phenomenon of benign overfitting, where a trained neural network perfectly fits noisy training data but still achieves near-optimal test performance, has been extensively studied in recent years for linear models and fully-connected/convolutional
Externí odkaz:
http://arxiv.org/abs/2410.07746
We study what provable privacy attacks can be shown on trained, 2-layer ReLU neural networks. We explore two types of attacks; data reconstruction attacks, and membership inference attacks. We prove that theoretical results on the implicit bias of 2-
Externí odkaz:
http://arxiv.org/abs/2410.07632
Autor:
Frei, Spencer, Vardi, Gal
Transformers have the capacity to act as supervised learning algorithms: by properly encoding a set of labeled training ("in-context") examples and an unlabeled test example into an input sequence of vectors of the same dimension, the forward pass of
Externí odkaz:
http://arxiv.org/abs/2410.01774