AS
Ashkan Soleymani
cs.LGstat.MLmath.OCcs.GTcs.AIcs.SYeess.SYstat.MECommunicable DiseasesCOVID-19
On Valency
published · living versionsW_h6dd646v·v1 · currentpublished
On Lipschitz Explosion in Deep Neural Networks with Normalization: Consequences for Optimization and Adversarial Robustness
with Reyhaneh Hosseinpourkhoshkbari, Hadi Daneshmand, Patrick Jaillet
1 version
Preprints & journals
20 papers in the corpus · 2020–2026Breaking the 1/3 Barrier for Approximate Nash Equilibria in Symmetric Bimatrix Games2610.25524v1 · Ashkan Soleymani, Patrick Jaillet, Gabriele Farina2026 · 0 citationsarXivon Valency
Multiplicative Optimism for Constant Regret in Games2609.21976v1 · Ashkan Soleymani, Georgios Piliouras2026 · 0 citationsarXiv
Exact-Form Regret for Gradient Descent, Mirror Descent and Follow-the-Regularized-Leader2609.09466v2 · Ashkan Soleymani, Gabriele Farina, Patrick Jaillet2026 · 0 citationsarXivon Valency
Efficient Learning and Symmetry Discovery under Exact Invariances2609.07031v1 · Ashkan Soleymani, Behrooz Tahmasebi, Patrick Jaillet et al.2026 · 0 citationsConference on Learning Theory (COLT) 2026
Algorithmic Optimality Guarantees for Nonsmooth Output-Feedback Policy Search2609.06197v1 · Ashkan Soleymani, Patrick Jaillet2026 · 0 citationsarXiv
Data Generation without Function Estimation2507.08239v1 · Hadi Daneshmand, Ashkan Soleymani2025 · 0 citationsarXiv
Double Machine Learning for Conditional Moment Restrictions: IV Regression, Proximal Causal Learning and Beyond2506.14950v2 · Daqian Shao, Ashkan Soleymani, Francesco Quinzan et al.2025 · 0 citationsarXiv
Cautious Optimism: A Meta-Algorithm for Near-Constant Regret in General Games2506.05005v2 · Ashkan Soleymani, Georgios Piliouras, Gabriele Farina2025 · 0 citationsarXiv
Faster Rates for No-Regret Learning in General Games via Cautious Optimism2503.24340v1 · Ashkan Soleymani, Georgios Piliouras, Gabriele Farina2025 · 1 citationarXiv
Learning with Exact Invariances in Polynomial Time2502.19758v1 · Ashkan Soleymani, Behrooz Tahmasebi, Stefanie Jegelka et al.2025 · 0 citationsarXiv
A Universal Class of Sharpness-Aware Minimization Algorithms2406.03682v2 · Behrooz Tahmasebi, Ashkan Soleymani, Dara Bahri et al.2024 · 0 citationsarXiv
Learning Decision Policies with Instrumental Variables through Double Machine Learning2405.08498v3 · Daqian Shao, Ashkan Soleymani, Francesco Quinzan et al.2024 · 0 citationsProceedings of the 41st International Conference on Machine Learning, PMLR 235:44489-44514, 2024
Double Machine Learning Based Structure Identification from Temporal Data2311.06012v2 · Emmanouil Angelis, Francesco Quinzan, Ashkan Soleymani et al.2023 · 0 citationsarXiv
DRCFS: Doubly Robust Causal Feature Selection2306.07024v3 · Francesco Quinzan, Ashkan Soleymani, Patrick Jaillet et al.2023 · 1 citationarXiv
Pyfectious: An individual-level simulator to discover optimal containment policies for epidemic diseases.36689461 · Mehrjou, Arash, Soleymani, Ashkan, Abyaneh, Amin et al.2023 · 3 citationsPLoS computational biology. 2023;19(1):e1010799
Federated Learning in Multi-Center Critical Care Research: A Systematic Case Study using the eICU Database2204.09328v1 · Arash Mehrjou, Ashkan Soleymani, Annika Buchholz et al.2022 · 2 citationsarXiv
Physical Derivatives: Computing policy gradients by physical forward-propagation2201.05830v1 · Arash Mehrjou, Ashkan Soleymani, Stefan Bauer et al.2022 · 0 citationsarXiv
GeneDisco: A Benchmark for Experimental Design in Drug Discovery2110.11875v1 · Arash Mehrjou, Ashkan Soleymani, Andrew Jesson et al.2021 · 8 citationsarXiv
Pyfectious: An individual-level simulator to discover optimal containment polices for epidemic diseases2103.15561v2 · Arash Mehrjou, Ashkan Soleymani, Amin Abyaneh et al.2021 · 1 citationarXiv
Causal Feature Selection via Orthogonal Search2007.02938v3 · Ashkan Soleymani, Anant Raj, Stefan Bauer et al.2020 · 4 citationsarXiv
Career total: 24 works. 20 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.