Arthur P. Feeneyhttps://arthurfeeney.github.io/Recent content on Arthur P. FeeneyHugo -- 0.156.0enTue, 10 Mar 2026 00:00:00 +0000NUCLEUS:https://arthurfeeney.github.io/papers/nucleus/Tue, 10 Mar 2026 00:00:00 +0000https://arthurfeeney.github.io/papers/nucleus/asdfAlloyMoE: GPU Kernels for Mixture of Expertshttps://arthurfeeney.github.io/papers/alloy-moe/Fri, 06 Mar 2026 00:00:00 +0000https://arthurfeeney.github.io/papers/alloy-moe/This project is looking at developing performant and memory efficient GPU kernels for MoE modules.BERN-NN-IBF: Enhancing Neural Network Bound Propagation Through Implicit Bernstein Form and Optimized Tensor Operationshttps://arthurfeeney.github.io/papers/bern-nn/Wed, 06 Nov 2024 00:00:00 +0000https://arthurfeeney.github.io/papers/bern-nn/This paper explores methods to perform efficient bounds-propagation on neural networks. Published in IEEE Transactions on Computer-Aided Design.BubbleML: A Multi-Physics Dataset and Benchmarks for Machine Learninghttps://arthurfeeney.github.io/papers/bubbleml/Sun, 10 Dec 2023 00:00:00 +0000https://arthurfeeney.github.io/papers/bubbleml/This paper creates a challenging multiphase, multiphysics dataset for PDE Surrogates and does analysis of current limitations. Spotlight paper at NeurIPS 2023Mosaic Flowshttps://arthurfeeney.github.io/papers/mosaic-flows/Tue, 14 Nov 2023 00:00:00 +0000https://arthurfeeney.github.io/papers/mosaic-flows/This paper extended Mosaic Flows to scale training and inference to distributed GPUs. This greatly improved the training time.