Logan Grosenick

Led by Marzieh Ajirak, with @NiliSolomonov and me co-senior and amazing @WeillCornell collaborators @OdedBein, Rose Bowen, Dora Kanellopoulos, @avitalfalk, and Faith Gunning.

A key result: significant error reduction for depression & anxiety prediction using unstructured and structured data, with explainable routing policies that match clinical intuition.

New preprint from the lab, accepted at @NeurIPSConf 2025! arxiv.org/pdf/2509.12227 Most AI models treat all inputs the same. We show how to adaptively route each sample across modalities and tasks, boosting performance, interpretability, and clinical impact for heterogenous data.

Come by the poster today! West Ballroom A-D #5301 11am-2pm

We hope this benchmark will facilitate more reliable assessment of reinforcement learning in dynamic healthcare settings and help accelerate the development of better RL algorithms and training protocols appropriate for medical applications.

The lab's third paper at @NeurIPSConf, first authored by Mason Hargrave, introduces a first benchmark for reinforcement learning in healthcare settings.

Come by the poster today at @NeurIPSConf ! East Exhibit Hall A-C #3000

This extends the successful local inductive bias of CNNs first shown to greatly enhance network performance by @ylecun to graphs.

Our second paper at NeuIPS, first authored by Isaac Osafo Nkansah, properly extends CNNs to graph-structured data. Our QuantNets capture complex patterns in irregular structures—like molecular graphs, fluid simulations, and neuroimaging—just as CNNs do for images. neurips.cc/virtual/2024/p…

Check out Amanda's poster tonight: WiML Poster 227 (Amanda Buch) / Poster board 5089 Tuesday, December 10, 6:30 pm – 8:30 pm Joint Poster Session for Affinity Workshops (East Ballroom AB + West Ballroom D) And check out the full paper here: proceedings.mlr.press/v238/buch24a/b…

The result? More transparent, data-driven patient subgroup identification from multiomics and imaging, improving the path to personalized treatments.

Very excited for my lab to be at @NeurIPSConf this year with three projects. The first, by very talented postdoc @amandambuch at @WiMLworkshop introduces a simple, scalable approach to jointly cluster and embed high-dimensional biomedical data.

🚨Big news! Five years in the making, our Zebrahub paper is now published in #Cell 🎉. We’ve built a timecourse atlas of zebrafish embryonic development, combining #scRNAseq data and #lightsheet live imaging, and delved deep into the dynamics of key progenitors. cell.com/cell/fulltext/… Explore the datasets here: zebrahub.sf.czbiohub.org Led by @Merlin_Lange with key contributions by @ale_agranados, @Shruthi94Vijay, @jobragantini, @Sarah_E_Ancheta and @yangjoonkim #zebrahub #devbio #cell #Biology #imaging @CellCellPress @czbbiohub #CZBiohubS