Showing 1–3 of 3 results for author: Woo, D

Iterated Energy-based Flow Matching for Sampling from Boltzmann Densities

Authors: Dongyeop Woo, Sungsoo Ahn

Abstract: In this work, we consider the problem of training a generator from evaluations of energy functions or unnormalized densities. This is a fundamental problem in probabilistic inference, which is crucial for scientific applications such as learning the 3D coordinate distribution of a molecule. To solve this problem, we propose iterated energy-based flow matching (iEFM), the first off-policy approach… ▽ More In this work, we consider the problem of training a generator from evaluations of energy functions or unnormalized densities. This is a fundamental problem in probabilistic inference, which is crucial for scientific applications such as learning the 3D coordinate distribution of a molecule. To solve this problem, we propose iterated energy-based flow matching (iEFM), the first off-policy approach to train continuous normalizing flow (CNF) models from unnormalized densities. We introduce the simulation-free energy-based flow matching objective, which trains the model to predict the Monte Carlo estimation of the marginal vector field constructed from known energy functions. Our framework is general and can be extended to variance-exploding (VE) and optimal transport (OT) conditional probability paths. We evaluate iEFM on a two-dimensional Gaussian mixture model (GMM) and an eight-dimensional four-particle double-well potential (DW-4) energy function. Our results demonstrate that iEFM outperforms existing methods, showcasing its potential for efficient and scalable probabilistic modeling in complex high-dimensional systems. △ Less

Submitted 29 August, 2024; originally announced August 2024.

Non-backtracking Graph Neural Networks

Authors: Seonghyun Park, Narae Ryu, Gahee Kim, Dongyeop Woo, Se-Young Yun, Sungsoo Ahn

Abstract: The celebrated message-passing updates for graph neural networks allow representing large-scale graphs with local and computationally tractable updates. However, the updates suffer from backtracking, i.e., a message flowing through the same edge twice and revisiting the previously visited node. Since the number of message flows increases exponentially with the number of updates, the redundancy in… ▽ More The celebrated message-passing updates for graph neural networks allow representing large-scale graphs with local and computationally tractable updates. However, the updates suffer from backtracking, i.e., a message flowing through the same edge twice and revisiting the previously visited node. Since the number of message flows increases exponentially with the number of updates, the redundancy in local updates prevents the graph neural network from accurately recognizing a particular message flow relevant for downstream tasks. In this work, we propose to resolve such a redundancy issue via the non-backtracking graph neural network (NBA-GNN) that updates a message without incorporating the message from the previously visited node. We theoretically investigate how NBA-GNN alleviates the over-squashing of GNNs, and establish a connection between NBA-GNN and the impressive performance of non-backtracking updates for stochastic block model recovery. Furthermore, we empirically verify the effectiveness of our NBA-GNN on the long-range graph benchmark and transductive node classification problems. △ Less

Submitted 25 September, 2024; v1 submitted 11 October, 2023; originally announced October 2023.

Differential impact of acute fine particulate matter exposure on risk of stroke by stroke subtype, age, sex, and race: a case-crossover study

Authors: Mohammad Alfrad Nobel Bhuiyan, Cole Brokamp, Tracy E. Madsen, Jane C. Khoury, Heidi Sucharew, Kathleen Alwell, Charles J. Moomaw, Monir Hossain, Matthew L. Flaherty, Daniel Woo, Jason Mackey, Felipe De Los Rios La Rosa, Sharyl Martini, Simona Ferioli, Opeolu Adeoye, Brett M. Kissela, Dawn Kleindorfer

Abstract: Objective: Association between acute ambient fine particulate matter (PM2.5, aerodynamic diameter less than 2.5 and cardiovascular events are well documented. However, it remains unclear whether acute exposure to PM2.5 acts as a trigger for hemorrhagic (intracerebral or subarachnoid hemorrhage) or non-hemorrhagic (infarct or transient ischemic attack) stroke onset. We, therefore, examined the asso… ▽ More Objective: Association between acute ambient fine particulate matter (PM2.5, aerodynamic diameter less than 2.5 and cardiovascular events are well documented. However, it remains unclear whether acute exposure to PM2.5 acts as a trigger for hemorrhagic (intracerebral or subarachnoid hemorrhage) or non-hemorrhagic (infarct or transient ischemic attack) stroke onset. We, therefore, examined the association between ambient PM2.5 and stroke onset, and whether this relationship differs by stroke subtype, age, sex, and race. Methods: We used a time-stratified case-crossover design to examine the association between exposure to PM2.5 and stroke onset for the calendar year 2010. Data were collected from the Greater Cincinnati Northern Kentucky Stroke Study. We included patients 20 years and older, initially ascertained through hospital ICD-9 discharge codes. Daily ambient concentrations of PM2.5 were estimated from the patient's residential addresses using a spatiotemporal model. Conclusion: Our findings suggest that short-term PM2.5 exposure, particularly at three days before the event, is associated with stroke onset but varies according to stroke subtype, age, sex, and race. △ Less

Submitted 31 December, 2019; v1 submitted 19 December, 2019; originally announced December 2019.

Comments: One coauthor didnt want to submit the paper on the arxiv