Neural Low-Degree Filtering (Neural LoFi) is introduced, a stylized limit of gradient-based training in which hierarchical feature learning becomes an explicit iterative spectral procedure and provides a mathematically explicit framework for studying multi-layer feature learning beyond the lazy regime.
Decision-makers rely on weather forecasts to plant crops, manage wildfires, allocate water and energy, and prepare for weather extremes. Today, such forecasts enjoy unprecedented accuracy out to two weeks thanks to steady advances in physics-based dynamical models and data-driven artificial intelligence (AI) models. Ho...
Hannah Guan, S. Mouatadid, Paulo Orenstein et al.· 1 citation
Forecasting time series accurately is critical for applications with complex data ranging from energy systems to healthcare and finance. Among current state of the art models, generative latent variable models are increasingly implemented; yet principled generalisation guarantees for modern latent variable models remai...
Chlo\'e Hashimoto-Cullen, Ghislain Agoua, Benjamin Guedj et al.· 0 citations
Dynamical symbolic regression methods identify governing differential equations from noisy data, balancing interpretability and predictive accuracy. However, standard methods often produce expressions that violate known physical laws. To address this, we propose FluxDisco, a physics-informed framework tailored for flux...
Cassandra Durr (Lancaster University), Alvaro K\"ohn-Luque (University of Oslo), Chris Jewell (Lancaster University) et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
It is shown that pooling augmentations together, despite inter-dependencies, is a better alternative than the baseline of partitioning the data into subsets of independent data and provides an intuition behind the practical preference towards using many augmentations.
Maximilian Fleissner, D. Ghoshdastidar, Samory Kpotufe· 0 citations
While diffusion-based methods have recently emerged as effective tools for probing the intrinsic geometry of high-dimensional data, their statistical difficulty remains largely unexplored. We study estimation of the finite-scale population functional underlying FLIPD (Kamkari et al., 2024; arXiv:2406.03537), a diffusio...
A constraint-aware conditional generative framework for synthetic origin-destination demand generation in hierarchical logistics networks, which models demand as a conditional distribution over destinations given each origin, enabling topology-aware synthesis that is both topologically realistic and operationally feasi...
LeJEPA is adapted to molecular graphs, evaluating GPS and Chemprop-style D-MPNN encoders on the antibiotic-activity dataset and ogbg-molhiv using a multi-seed, bootstrap-based protocol, and pretraining supplies complementary information best realised through feature-level combination, while finetuning gains are weak an...
Michał Kulczykowski, Rafał Łabędzki· Research Square· 0 citations
Late Gadolinium Enhancement (LGE) on cardiac magnetic resonance is a key marker of myocardial scar, but its limited accessibility motivates routine ECG-based screening. We evaluated whether $\beta$-variational autoencoder (VAE)-derived ECG representations can discriminate LGE+ from LGE- cardiomyopathic patients in a lo...
Shayan Sharifi, R. Treu, I. Gandin et al.· 0 citations
Experiments on two representative radio access network (RAN) control tasks show that CV-CCI remains valid under hidden confounding while producing more efficient prediction sets than state-of-the-art confounding-valid baselines.
We show that susceptibilities, an interpretability technique developed for neural networks, can identify the presence of algorithmic structure in Turing machines by probing the local loss landscape of a learning problem for noisy Turing machines introduced by Murfet and Troiani (arXiv:2504.08075). We prove that symmetr...
Billy Snikkers, Rumi Salazar, Daniel Murfet et al.· 0 citations
Nuclear fusion plays a pivotal role in the quest for reliable and sustainable energy production. A major roadblock to viable fusion power is understanding plasma turbulence, which significantly impairs plasma confinement, and is vital for next-generation reactor design. Plasma turbulence is governed by the nonlinear gy...
Fabian Paischer, G. Galletti, W. Hornsby et al.· Neural Information Processin...· 9 citations
A machine-learning approach called CheMeleon provides a practical way to develop new drugs or other molecules with much less detailed, expensive experimentation.
