Publications related to Laboratory for Information and Inference Systems

Regularization Techniques for Low-Resource Machine Translation

Neural machine translation (MT) and text generation have recently reached very high levels of quality. However, both areas share a problem: in order to reach these levels, they require massive amounts of data. When this is not present, they lack generaliza ...

EPFL2023

BLACK-BOX ATTACKS ON IMAGE ACTIVITY PREDICTION AND ITS NATURAL LANGUAGE EXPLANATIONS

Andrea Cavallaro

Explainable AI (XAI) methods aim to describe the decision process of deep neural networks. Early XAI methods produced visual explanations, whereas more recent techniques generate multimodal explanations that include textual information and visual represent ...

Los Alamitos2023

Leveraging Unlabeled Data to Track Memorization

Patrick Thiran, Mahsa Forouzesh, Hanie Sedghi

Deep neural networks may easily memorize noisy labels present in real-world data, which degrades their ability to generalize. It is therefore important to track and evaluate the robustness of models against noisy label memorization. We propose a metric, ca ...

2023

Towards Stable and Efficient Adversarial Training against $l_1$ Bounded Adversarial Attacks

Sabine Süsstrunk, Mathieu Salzmann, Yulun Jiang, Chen Liu, Zhuoyi Huang

We address the problem of stably and efficiently training a deep neural network robust to adversarial perturbations bounded by an

l_1

norm. We demonstrate that achieving robustness against

l_1

-bounded perturbations is more challenging than in the

l_2

...

2023

Penalising the biases in norm regularisation enforces sparsity

Nicolas Henri Bernard Flammarion, Etienne Patrice Boursier

Controlling the parameters' norm often yields good generalisation when training neural networks. Beyond simple intuitions, the relation between parameters' norm and obtained estimators theoretically remains misunderstood. For one hidden ReLU layer networks ...

2023

Can Self-Supervised Neural Networks Pre-Trained on Human Speech distinguish Animal Callers?

Mathew Magimai Doss, Eklavya Sarkar

Self-supervised learning (SSL) models use only the intrinsic structure of a given signal, independent of its acoustic domain, to extract essential information from the input to an embedding space. This implies that the utility of such representations is no ...

ISCA2023

Safe Zeroth-Order Convex Optimization Using Quadratic Local Approximations

Giancarlo Ferrari Trecate, Maryam Kamgarpour, Yuning Jiang, Baiwei Guo

We address black-box convex optimization problems, where the objective and constraint functions are not explicitly known but can be sampled within the feasible set. The challenge is thus to generate a sequence of feasible points converging towards an optim ...

2023

A new variable shape parameter strategy for RBF approximation using neural networks

Jan Sickmann Hesthaven

The choice of the shape parameter highly effects the behaviour of radial basis function (RBF) approximations, as it needs to be selected to balance between the ill-conditioning of the interpolation matrix and high accuracy. In this paper, we demonstrate ho ...

PERGAMON-ELSEVIER SCIENCE LTD2023

Controlled physics-informed data generation for deep learning-based remaining useful life prediction under unseen operation conditions

Olga Fink, Jian Zhou

Limited availability of representative time-to-failure (TTF) trajectories either limits the performance of deep learning (DL)-based approaches on remaining useful life (RUL) prediction in practice or even precludes their application. Generating synthetic d ...

2023

Fast Adversarial Training With Adaptive Step Size

Sabine Süsstrunk, Mathieu Salzmann, Chen Liu, Zhuoyi Huang, Yong Zhang, Jue Wang

While adversarial training and its variants have shown to be the most effective algorithms to defend against adversarial attacks, their extremely slow training process makes it hard to scale to large datasets like ImageNet. The key idea of recent works to ...

Piscataway2023

Laboratory for Information and Inference Systems

Graph Chatbot

Chat with Graph Search

Regularization Techniques for Low-Resource Machine Translation

BLACK-BOX ATTACKS ON IMAGE ACTIVITY PREDICTION AND ITS NATURAL LANGUAGE EXPLANATIONS

Leveraging Unlabeled Data to Track Memorization

Towards Stable and Efficient Adversarial Training against $l_1$ Bounded Adversarial Attacks

Penalising the biases in norm regularisation enforces sparsity

Can Self-Supervised Neural Networks Pre-Trained on Human Speech distinguish Animal Callers?

Safe Zeroth-Order Convex Optimization Using Quadratic Local Approximations

A new variable shape parameter strategy for RBF approximation using neural networks

Controlled physics-informed data generation for deep learning-based remaining useful life prediction under unseen operation conditions

Fast Adversarial Training With Adaptive Step Size

BLACK-BOX ATTACKS ON IMAGE ACTIVITY PREDICTION AND ITS NATURAL LANGUAGE EXPLANATIONS

Penalising the biases in norm regularisation enforces sparsity

A new variable shape parameter strategy for RBF approximation using neural networks

Controlled physics-informed data generation for deep learning-based remaining useful life prediction under unseen operation conditions

Fast Adversarial Training With Adaptive Step Size

Regularization Techniques for Low-Resource Machine Translation

Leveraging Unlabeled Data to Track Memorization

Towards Stable and Efficient Adversarial Training against $l_1$ Bounded Adversarial Attacks

Safe Zeroth-Order Convex Optimization Using Quadratic Local Approximations

Can Self-Supervised Neural Networks Pre-Trained on Human Speech distinguish Animal Callers?