Author name cluster

Waleed Mustafa

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

7 papers

2 author rows

AAAI Conference 2026 Conference Paper

TORA: Train Once, Realign Anytime for Offline Multi-Objective Reinforcement Learning

Weichen Li
Waleed Mustafa
Marcio Monteiro
Puyu Wang
Marius Kloft
Sophie Fellenz

Intelligent agents in real-world applications must adapt their behavior to changing contexts and user preferences. For example, planning a road trip requires considering both travel time and cost. Multi-objective reinforcement learning (MORL) provides a principled approach to navigate such trade-offs. However, most existing approaches require predefined preference weights during training and jointly optimize the model for all objectives. In this paper, we introduce TORA (Train Once, Realign Anytime), a novel framework that defers preference integration to inference time, enabling flexible adaptation to user preferences without retraining. TORA independently trains diffusion planning models for each objective and combines them at inference time using user-specified preferences to generate behavior aligned with desired trade-offs. Furthermore, new objectives can be added seamlessly by training additional models without modifying existing ones. Empirical evaluations on standard offline MORL benchmarks demonstrate that TORA achieves competitive and consistent performance compared to methods that require fixed preference weights.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Mitigating Spurious Features in Contrastive Learning with Spectral Regularization

Naghmeh Ghanooni
Waleed Mustafa
Dennis Wagner
Sophie Fellenz
Anthony Lin
Marius Kloft

Neural networks generally prefer simple and easy-to-learn features. When these features are spuriously correlated with the labels, the network's performance can suffer, particularly for underrepresented classes or concepts. Self-supervised representation learning methods, such as contrastive learning, are especially prone to this issue, often resulting in worse performance on downstream tasks. We identify a key spectral signature of this failure: early reliance on dominant singular modes of the learned feature matrix. To mitigate this, we propose a novel framework that promotes a uniform eigenspectrum of the feature covariance matrix, encouraging diverse and semantically rich representations. Our method operates in a fully self-supervised setting, without relying on ground-truth labels or any additional information. Empirical results on SimCLR and SimSiam demonstrate consistent gains in robustness and transfer performance, suggesting broad applicability across self-supervised learning paradigms. Code: https: //github. com/NaghmehGh/SpuriousCorrelation_SSRL

PDF Details

TMLR Journal 2024 Journal Article

Generalization Bounds with Logarithmic Negative-Sample Dependence for Adversarial Contrastive Learning

Naghmeh Ghanooni
Waleed Mustafa
Yunwen Lei
Anthony Widjaja Lin
Marius Kloft

Contrastive learning has emerged as a powerful unsupervised learning technique for extracting meaningful representations from unlabeled data by pulling similar data points closer in the representation space and pushing dissimilar ones apart. However, its vulnerability to adversarial attacks remains a critical challenge. To address this, adversarial contrastive learning — incorporating adversarial training into contrastive loss — has emerged as a promising approach to achieving robust representations that can withstand various adversarial attacks. While empirical evidence highlights its effectiveness, a comprehensive theoretical framework has been lacking. In this paper, we fill this gap by introducing generalization bounds for adversarial contrastive learning, offering key theoretical insights. Leveraging the Lipschitz continuity of loss functions, we derive generalization bounds that scale logarithmically with the number of negative samples, $K$, and apply to both linear and non-linear representations, including those obtained from deep neural networks (DNNs). Our theoretical results are supported by experiments on real-world datasets.

PDF Details

IJCAI Conference 2024 Conference Paper

Interpretable Tensor Fusion

Saurabh Varshneya
Antoine Ledent
Philipp Liznerski
Andriy Balinskyy
Purvanshi Mehta
Waleed Mustafa
Marius Kloft

Conventional machine learning methods are predominantly designed to predict outcomes based on a single data type. However, practical applications may encompass data of diverse types, such as text, images, and audio. We introduce interpretable tensor fusion (InTense), a multimodal learning method training a neural network to simultaneously learn multiple data representations and their interpretable fusion. InTense can separately capture both linear combinations and multiplicative interactions of the data types, thereby disentangling higher-order interactions from the individual effects of each modality. InTense provides interpretability out of the box by assigning relevance scores to modalities and their associations, respectively. The approach is theoretically grounded and yields meaningful relevance scores on multiple synthetic and real-world datasets. Experiments on four real-world datasets show that InTense outperforms existing state-of-the-art multimodal interpretable approaches in terms of accuracy and interpretability.

PDF Details DOI

ICML Conference 2022 Conference Paper

On the Generalization Analysis of Adversarial Learning

Waleed Mustafa
Yunwen Lei
Marius Kloft

Many recent studies have highlighted the susceptibility of virtually all machine-learning models to adversarial attacks. Adversarial attacks are imperceptible changes to an input example of a given prediction model. Such changes are carefully designed to alter the otherwise correct prediction of the model. In this paper, we study the generalization properties of adversarial learning. In particular, we derive high-probability generalization bounds on the adversarial risk in terms of the empirical adversarial risk, the complexity of the function class and the adversarial noise set. Our bounds are generally applicable to many models, losses, and adversaries. We showcase its applicability by deriving adversarial generalization bounds for the multi-class classification setting and various prediction models (including linear models and Deep Neural Networks). We also derive optimistic adversarial generalization bounds for the case of smooth losses. These are the first fast-rate bounds valid for adversarial deep learning to the best of our knowledge.

Details

IJCAI Conference 2021 Conference Paper

Fine-grained Generalization Analysis of Structured Output Prediction

Waleed Mustafa
Yunwen Lei
Antoine Ledent
Marius Kloft

In machine learning we often encounter structured output prediction problems (SOPPs), i. e. problems where the output space admits a rich internal structure. Application domains where SOPPs naturally occur include natural language processing, speech recognition, and computer vision. Typical SOPPs have an extremely large label set, which grows exponentially as a function of the size of the output. Existing generalization analysis implies generalization bounds with at least a square-root dependency on the cardinality d of the label set, which can be vacuous in practice. In this paper, we significantly improve the state of the art by developing novel high-probability bounds with a logarithmic dependency on d. Furthermore, we leverage the lens of algorithmic stability to develop generalization bounds in expectation without any dependency on d. Our results therefore build a solid theoretical foundation for learning in large-scale SOPPs. Furthermore, we extend our results to learning with weakly dependent data.

PDF Details DOI

AAAI Conference 2021 Conference Paper

Norm-Based Generalisation Bounds for Deep Multi-Class Convolutional Neural Networks

Antoine Ledent
Waleed Mustafa
Yunwen Lei
Marius Kloft

We show generalisation error bounds for deep learning with two main improvements over the state of the art. (1) Our bounds have no explicit dependence on the number of classes except for logarithmic factors. This holds even when formulating the bounds in terms of the Frobenius-norm of the weight matrices, where previous bounds exhibit at least a squareroot dependence on the number of classes. (2) We adapt the classic Rademacher analysis of DNNs to incorporate weight sharing—a task of fundamental theoretical importance which was previously attempted only under very restrictive assumptions. In our results, each convolutional filter contributes only once to the bound, regardless of how many times it is applied. Further improvements exploiting pooling and sparse connections are provided. The presented bounds scale as the norms of the parameter matrices, rather than the number of parameters. In particular, contrary to bounds based on parameter counting, they are asymptotically tight (up to log factors) when the weights approach initialisation, making them suitable as a basic ingredient in bounds sensitive to the optimisation procedure. We also show how to adapt the recent technique of loss function augmentation to replace spectral norms by empirical analogues whilst maintaining the advantages of our approach.

PDF Details