Jules J. Berman
YOU?
Author Swipe
View article: DICE: Discrete inverse continuity equation for learning population dynamics
DICE: Discrete inverse continuity equation for learning population dynamics Open
We introduce the Discrete Inverse Continuity Equation (DICE) method, a generative modeling approach that learns the evolution of a stochastic process from given sample populations at a finite number of time points. Models learned with DICE…
View article: JSONSchemaBench: A Rigorous Benchmark of Structured Outputs for Language Models
JSONSchemaBench: A Rigorous Benchmark of Structured Outputs for Language Models Open
Reliably generating structured outputs has become a critical capability for modern language model (LM) applications. Constrained decoding has emerged as the dominant technology across sectors for enforcing structured outputs during generat…
View article: Parametric model reduction of mean-field and stochastic systems via higher-order action matching
Parametric model reduction of mean-field and stochastic systems via higher-order action matching Open
The aim of this work is to learn models of population dynamics of physical systems that feature stochastic and mean-field effects and that depend on physics parameters. The learned models can act as surrogates of classical numerical models…
View article: CoLoRA: Continuous low-rank adaptation for reduced implicit neural modeling of parameterized partial differential equations
CoLoRA: Continuous low-rank adaptation for reduced implicit neural modeling of parameterized partial differential equations Open
This work introduces reduced models based on Continuous Low Rank Adaptation (CoLoRA) that pre-train neural networks for a given partial differential equation and then continuously adapt low-rank weights in time to rapidly predict the evolu…
View article: Neuronal temporal filters as normal mode extractors
Neuronal temporal filters as normal mode extractors Open
Physical Review Research To generate actions in the face of physiological delays, the brain must predict the future. Here we explore how prediction may lie at the core of brain function by considering a neuron predicting the future of a sc…
View article: Neuronal Temporal Filters as Normal Mode Extractors
Neuronal Temporal Filters as Normal Mode Extractors Open
To generate actions in the face of physiological delays, the brain must predict the future. Here we explore how prediction may lie at the core of brain function by considering a neuron predicting the future of a scalar time series input. A…
View article: Nonlinear embeddings for conserving Hamiltonians and other quantities with Neural Galerkin schemes
Nonlinear embeddings for conserving Hamiltonians and other quantities with Neural Galerkin schemes Open
This work focuses on the conservation of quantities such as Hamiltonians, mass, and momentum when solution fields of partial differential equations are approximated with nonlinear parametrizations such as deep networks. The proposed approa…
View article: Randomized Sparse Neural Galerkin Schemes for Solving Evolution Equations with Deep Networks
Randomized Sparse Neural Galerkin Schemes for Solving Evolution Equations with Deep Networks Open
Training neural networks sequentially in time to approximate solution fields of time-dependent partial differential equations can be beneficial for preserving causality and other physics properties; however, the sequential-in-time training…
View article: Representational dissimilarity metric spaces for stochastic neural networks
Representational dissimilarity metric spaces for stochastic neural networks Open
Quantifying similarity between neural representations -- e.g. hidden layer activation vectors -- is a perennial problem in deep learning and neuroscience research. Existing methods compare deterministic responses (e.g. artificial networks …
View article: Bridging the Gap: Point Clouds for Merging Neurons in Connectomics
Bridging the Gap: Point Clouds for Merging Neurons in Connectomics Open
In the field of Connectomics, a primary problem is that of 3D neuron segmentation. Although deep learning-based methods have achieved remarkable accuracy, errors still exist, especially in regions with image defects. One common type of def…
View article: Finding relationships among biological entities
Finding relationships among biological entities Open
View article: Changing how we think about infectious diseases
Changing how we think about infectious diseases Open
View article: Viruses
Viruses Open