2025
NeurIPS COML
workshop
Causal-Informed Hybrid Online Adaptive Optimization for Ad-Load Personalization in Large-Scale Social Networks
NeurIPS 2025 Workshop on Constrained Optimization for Machine Learning (COML).
/ research
Pushing on the
edges of what
a model can do.
A working list of papers, ablations and small experiments. The thread that ties them together is the same one I've been pulling on since grad school: how to make computation behave more like the biology that inspired it — efficient, robust, generative.
5
peer-reviewed papers
4
research arenas
9
open implementations
13yr
since first publication
filter:
N° 01 / 04
Causal Inference treatment effects, sequential randomization, ad-load policy
2025
arXiv · preprint
Treatment Effect Learning Under Sequential Randomization
arXiv preprint, 2025.
N° 02 / 04
Recommendation Systems scaling, retrieval, slate optimization
2022
KDD · proc.
first author
NxtPost: User-to-Post Recommendations in Facebook Groups
Proc. ACM SIGKDD (KDD), 2022, pp. 3792–3800.
N° 03 / 04
Computer Vision shape-prior segmentation, template networks
2019
IEEE TMI
journal
TeTrIS: Template Transformer Networks for Image Segmentation With Shape Priors
IEEE Transactions on Medical Imaging (TMI), 38(11):2596–2606.
N° 04 / 04
Robotics localization, state estimation
2013
IEEE ICACC
conference
Localization of a Mobile Autonomous Robot Using Extended Kalman Filter
IEEE International Conference on Advances in Computing, Communication and Control (ICACC), 2013.
Long-form notes that don't (yet) belong in a paper live next door, in /notes →