A. MARTINA NEUMAN

Website update (May 31 2026)!!! The CV and publication pages were recently repaired after an incorrect rendering issue. If you encounter any display problems, please let me know. 

Interdisciplinary mathematician - theorist by training. Ph.D.,UC Berkeley. Ph.D., UC Berkeley; postdoctoral researcher at the University of Vienna. 

US citizen residing in Austria.

neumana53 [at] univie [dot] ac [dot] at; marsneuman [dot] work [dot] gmail

Social media (gradually active): LinkedIn (same name)

What's new (short list)

Works and publications

• My paper exploring the theoretical foundation of contrastive learning is on route to get accepted at Journal of Machine Learning Research

• My paper exploring  learning guarantees through metric embedding is published with IEEE Transactions of Information Theory

• My paper  on spiking neural networks is published with Journal of Machine Learning Research

• My paper on practical existence theorem (PET) using Slepian basis is on route to get accepted at Information and Inference: a Journal of the IMA

• My paper on theoretical advantage of GNNs is published with Information and Inference: a Journal of the IMA

• A new paper on in-context vs agentic learning under realizability constraints

• A new paper on probing internal looped reasoning dynamics

• A new paper on transductive learning guarantees and small random worlds

(Expecting soon)

• A new paper on consistency of graph-based algorithms

• A new paper on spiking neural networks

• A new paper on fast transforms of SSLOs

Talks

• I will present a poster on probing internal looped reasoning dynamics at FoCM26

• I will present at SIAM DS 26

• I presented consistency of augmentation graph Laplacian and contrastive learning at SampTA 25

Current collaboration network