Aurélien Lucchi

University of Basel — Switzerland

Computer Science · Physical Sciences

28h-index12.5kcitations154works0.22yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Stochastic Gradient Optimization Techniques(42), Sparse and Compressive Sensing Techniques(25), Neural Networks and Applications(21), Generative Adversarial Networks and Image Synthesis(17), Model Reduction and Neural Networks(15)

Publications154 total

Optimizer choice matters for the emergence of Neural Collapse
ArXiv.org·2026Open Access
Optimizer choice matters for the emergence of Neural Collapse
arXiv (Cornell University)·2026Open Access
Multi-Task GRPO: Reliable LLM Reasoning Across Tasks
arXiv (Cornell University)·2026Open Access
Multi-Task GRPO: Reliable LLM Reasoning Across Tasks
ArXiv.org·2026Open Access
Synchronizing pretrained kernel regressors with applications to American option pricing
Frontiers of Mathematical Finance·2026Open Access
Noise-Induced Equalization in quantum learning models
arXiv (Cornell University)·2025Open Access
Why Do We Need Warm-up? A Theoretical Perspective
arXiv (Cornell University)·2025Open Access
Enhancing Optimizer Stability: Momentum Adaptation of The NGN Step-size
arXiv (Cornell University)·2025Open Access
Optimization Guarantees for Square-Root Natural-Gradient Variational Inference
arXiv (Cornell University)·2025Open Access
Trainability of Quantum Models Beyond Known Classical Simulability
arXiv (Cornell University)·2025Open Access
Sharp Generalization Bounds for Foundation Models with Asymmetric Randomized Low-Rank Adapters
arXiv (Cornell University)·2025Open Access
Unpacking Softmax: How Temperature Drives Representation Collapse, Compression, and Generalization
arXiv (Cornell University)·2025Open Access
Where You Place the Norm Matters: From Prejudiced to Neutral Initializations
arXiv (Cornell University)·2025Open Access
Unbiased and Sign Compression in Distributed Learning: Comparing Noise Resilience via SDEs
arXiv (Cornell University)·2025Open Access
Double Momentum and Error Feedback for Clipping with Fast Rates and Differential Privacy
arXiv (Cornell University)·2025Open Access
Adaptive Methods through the Lens of SDEs: Theoretical Insights on the Role of Noise
arXiv (Cornell University)·2024Open Access
Theoretical characterisation of the Gauss-Newton conditioning in Neural Networks
arXiv (Cornell University)·2024Open Access
A Comprehensive Analysis on the Learning Curve in Kernel Ridge\n Regression
arXiv (Cornell University)·2024Open Access
Loss Landscape Characterization of Neural Networks without Over-Parametrization
arXiv (Cornell University)·2024Open Access
Cubic regularized subspace Newton for non-convex optimization
arXiv (Cornell University)·2024· 1 citedOpen Access
page 1 of 8Next →

Frequent Co-authors

Thomas Hofmann(28), Pascal Fua(9), Hadi Daneshmand(9), Jonas Köhler(7), Kevin Smith(6), Alexandre Réfrégier(6), Janis Fluri(5), A. Amara(4), T. Kacprzak(4), Radhakrishna Achanta(3), Christa Zoufal(3), Stefan Woerner(3), Kevin A. Roth(3), Sebastian Nowozin(3), Martin Jaggi(3), Klaus Neymeyr(3), Graham Knott(2), Zuoyue Li(2), Jan Dirk Wegner(2), Tomasz Kacprzak(2)