Many scientific and machine learning systems, from molecular dynamics to diffusion models and beyond, are governed by stochastic dynamics with low-dimensional structure, evolving on slow timescales. However, target trajectories, used to identify and interpret such dynamics, are often inaccessible: only biased or static...
Vladimir R. Kostic, Karim Lounici, Hélène Halconruy et al.· 0 citations
We study logistic regression on linearly separable data under gradient descent with a large constant stepsize $\eta$. Such dynamics may exhibit a characteristic Edge of Stability phenomenon, in which the loss initially oscillates before transitioning to a stable phase of monotone decrease. Existing work provides a tigh...
We introduce the Clifford Sheaf Neural Network (CSNN), an equivariant sheaf neural network for geometric graphs that places a Clifford algebra on each stalk of a cellular sheaf and transports multivector features along edges. The canonical choice of restriction map for sheaves with algebra-valued stalks is algebra homo...
Bayesian Optimisation (BO) is a powerful framework for the optimisation of expensive black-box functions, but typically requires refitting a surrogate and maximising an acquisition function at every evaluation step. In-context approaches based on Prior-data Fitted Networks (PFNs) amortise part of this cost by pre-train...
Luca Geminiani, Nadja Klein· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Distinct pre-trained models specialized for the same task converge to closely similar behavior, yet the parameter updates that produce it share no common coordinate system. Weight-space task arithmetic is therefore confined to a single model, and transporting an update between models requires a structural correspondenc...
Experiments show the proposed multi-task IRL method is robust to limited coverage, recovers rewards on and off of each task's support, and transfers to target environments at lower regret than baselines, with its computational advantage over per-task methods widening as tasks grow.
Allen Tran, Jia Wan, Nathan Kallus et al.· 0 citations
Recent theoretical work identified fundamental properties of representation geometry that shape inference ability of deep neural networks. These include signal-noise factorization (SNF), the ability to segregate signal from noise, and signal-signal factorization (SSF), the ability to segregate task-specific and task-ir...
We introduce Grand Canonical Generators (GCG), a generative framework that extends Boltzmann generators to the grand canonical ensemble. We present two designs. The first conditions a variable-size generative model on the chemical potential, sampling particle number and configuration jointly. The second factorizes the...
Andreas Burger, Malte Franke, Luka Mucko et al.· 0 citations
An exact, information-theoretic analysis of stochastic gradient descent and its variants is given, showing that a preconditioned SGD step is the posterior-mean update of a Gaussian Bayes model, and that its one-step regret splits into an intrinsic-time cost and a change in comparator information.
We introduce IrekoGPT, a post-hoc method for converting pretrained LLMs into slimmable models whose width can be adjusted at inference time. Building on SliceGPT, we retain its projection matrices without pruning them, allowing a single model to expose nested subnetworks at different widths. We improve robustness by ca...
Pietro Moriello, Pietro Buzzega, Angelo Porrello et al.· 0 citations
Contemporary Genomic Foundation Models (GFMs) rely on a DNA-as-a-string paradigm that employs masked token prediction objectives for pretraining. However, this abstraction does not explicitly model the biochemical, structural, and physical properties essential to biological function. Many molecular properties can be es...
Jeremy Levy, Ariel Larey, Yury Nahshan et al.· 0 citations
Time-series foundation models (TSFMs) perform strongly across forecasting tasks, but their per-series inference is ill-suited to neighbor-rich forecasting, where each query has access to related but nonidentical historical series. Continuous glucose monitoring (CGM) and Web/cloud workloads exemplify this setting: CGM t...
Hao-Nan Shi, Tong Wu, Chen-Cong Sun et al.· 0 citations
A machine-learning approach called CheMeleon provides a practical way to develop new drugs or other molecules with much less detailed, expensive experimentation.
Is your company’s AI sovereign? A practical guide for CIOs on AI vendor lock-in, EU Data Act switching, cloud sovereignty, and testing your exit plan before you need it. The post Sovereignty Is the Option to Leave appeared first on GPT-Lab.
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
