Yi-An Ma
YOU?
Author Swipe
View article: Zephyrus: An Agentic Framework for Weather Science
Zephyrus: An Agentic Framework for Weather Science Open
Foundation models for weather science are pre-trained on vast amounts of structured numerical data and outperform traditional weather forecasting systems. However, these models lack language-based reasoning capabilities, limiting their uti…
View article: Almost Linear Convergence under Minimal Score Assumptions: Quantized Transition Diffusion
Almost Linear Convergence under Minimal Score Assumptions: Quantized Transition Diffusion Open
Continuous diffusion models have demonstrated remarkable performance in data generation across various domains, yet their efficiency remains constrained by two critical limitations: (1) the local adjacency structure of the forward Markov p…
View article: Nearly Dimension-Independent Convergence of Mean-Field Black-Box Variational Inference
Nearly Dimension-Independent Convergence of Mean-Field Black-Box Variational Inference Open
We prove that, given a mean-field location-scale variational family, black-box variational inference (BBVI) with the reparametrization gradient converges at a rate that is nearly independent of explicit dimension dependence. Specifically, …
View article: Multi-Step Consistency Models: Fast Generation with Theoretical Guarantees
Multi-Step Consistency Models: Fast Generation with Theoretical Guarantees Open
Consistency models have recently emerged as a compelling alternative to traditional SDE-based diffusion models. They offer a significant acceleration in generation by producing high-quality samples in very few steps. Despite their empirica…
View article: seeBias: A Comprehensive Tool for Assessing and Visualizing AI Fairness
seeBias: A Comprehensive Tool for Assessing and Visualizing AI Fairness Open
Fairness in artificial intelligence (AI) prediction models is increasingly emphasized to support responsible adoption in high-stakes domains such as health care and criminal justice. Guidelines and implementation frameworks highlight the i…
View article: Purifying Approximate Differential Privacy with Randomized Post-processing
Purifying Approximate Differential Privacy with Randomized Post-processing Open
We propose a framework to convert $(\varepsilon, δ)$-approximate Differential Privacy (DP) mechanisms into $(\varepsilon', 0)$-pure DP mechanisms under certain conditions, a process we call ``purification.'' This algorithmic technique leve…
View article: Discovering Latent Causal Graphs from Spatiotemporal Data
Discovering Latent Causal Graphs from Spatiotemporal Data Open
Many important phenomena in scientific fields like climate, neuroscience, and epidemiology are naturally represented as spatiotemporal gridded data with complex interactions. Inferring causal relationships from these data is a challenging …
View article: ClimaQA: An Automated Evaluation Framework for Climate Question Answering Models
ClimaQA: An Automated Evaluation Framework for Climate Question Answering Models Open
The use of Large Language Models (LLMs) in climate science has recently gained significant attention. However, a critical issue remains: the lack of a comprehensive evaluation framework capable of assessing the quality and scientific valid…
View article: Log-concave Sampling from a Convex Body with a Barrier: a Robust and Unified Dikin Walk
Log-concave Sampling from a Convex Body with a Barrier: a Robust and Unified Dikin Walk Open
We consider the problem of sampling from a $d$-dimensional log-concave distribution $π(θ) \propto \exp(-f(θ))$ for $L$-Lipschitz $f$, constrained to a convex body with an efficiently computable self-concordant barrier function, contained i…
View article: A Skewness-Based Criterion for Addressing Heteroscedastic Noise in Causal Discovery
A Skewness-Based Criterion for Addressing Heteroscedastic Noise in Causal Discovery Open
Real-world data often violates the equal-variance assumption (homoscedasticity), making it essential to account for heteroscedastic noise in causal discovery. In this work, we explore heteroscedastic symmetric noise models (HSNMs), where t…
View article: Diffusion-BBO: Diffusion-Based Inverse Modeling for Online Black-Box Optimization
Diffusion-BBO: Diffusion-Based Inverse Modeling for Online Black-Box Optimization Open
Online black-box optimization (BBO) aims to optimize an objective function by iteratively querying a black-box oracle in a sample-efficient way. While prior studies focus on forward approaches such as Gaussian Processes (GPs) to learn a su…
View article: Accuracy on the wrong line: On the pitfalls of noisy data for out-of-distribution generalisation
Accuracy on the wrong line: On the pitfalls of noisy data for out-of-distribution generalisation Open
"Accuracy-on-the-line" is a widely observed phenomenon in machine learning, where a model's accuracy on in-distribution (ID) and out-of-distribution (OOD) data is positively correlated across different hyperparameters and data configuratio…
View article: Demystifying SGD with Doubly Stochastic Gradients
Demystifying SGD with Doubly Stochastic Gradients Open
Optimization objectives in the form of a sum of intractable expectations are rising in importance (e.g., diffusion models, variational autoencoders, and many more), a setting also known as "finite sum with infinite data." For these problem…
View article: Faster Sampling via Stochastic Gradient Proximal Sampler
Faster Sampling via Stochastic Gradient Proximal Sampler Open
Stochastic gradients have been widely integrated into Langevin-based methods to improve their scalability and efficiency in solving large-scale sampling problems. However, the proximal sampler, which exhibits much faster convergence than L…
View article: Reverse Transition Kernel: A Flexible Framework to Accelerate Diffusion Inference
Reverse Transition Kernel: A Flexible Framework to Accelerate Diffusion Inference Open
To generate data from trained diffusion models, most inference algorithms, such as DDPM, DDIM, and other variants, rely on discretizing the reverse SDEs or their equivalent ODEs. In this paper, we view such approaches as decomposing the en…
View article: Multi-Fidelity Residual Neural Processes for Scalable Surrogate Modeling
Multi-Fidelity Residual Neural Processes for Scalable Surrogate Modeling Open
Multi-fidelity surrogate modeling aims to learn an accurate surrogate at the highest fidelity level by combining data from multiple sources. Traditional methods relying on Gaussian processes can hardly scale to high-dimensional data. Deep …
View article: Diffusion Models as Constrained Samplers for Optimization with Unknown Constraints
Diffusion Models as Constrained Samplers for Optimization with Unknown Constraints Open
Addressing real-world optimization problems becomes particularly challenging when analytic objective functions or constraints are unavailable. While numerous studies have addressed the issue of unknown objectives, limited research has focu…
View article: Learning Granger Causality from Instance-wise Self-attentive Hawkes Processes
Learning Granger Causality from Instance-wise Self-attentive Hawkes Processes Open
We address the problem of learning Granger causality from asynchronous, interdependent, multi-type event sequences. In particular, we are interested in discovering instance-level causal structures in an unsupervised manner. Instance-level …
View article: Faster Sampling without Isoperimetry via Diffusion-based Monte Carlo
Faster Sampling without Isoperimetry via Diffusion-based Monte Carlo Open
To sample from a general target distribution $p_*\propto e^{-f_*}$ beyond the isoperimetric condition, Huang et al. (2023) proposed to perform sampling through reverse diffusion, giving rise to Diffusion-based Monte Carlo (DMC). Specifical…
View article: A Gradient-Based Optimization Method Using the Koopman Operator
A Gradient-Based Optimization Method Using the Koopman Operator Open
In this paper, we propose a novel approach to solving optimization problems by reformulating the optimization problem into a dynamical system, followed by the adaptive spectral Koopman (ASK) method. The Koopman operator, employed in our ap…
View article: Posterior Sampling with Delayed Feedback for Reinforcement Learning with Linear Function Approximation
Posterior Sampling with Delayed Feedback for Reinforcement Learning with Linear Function Approximation Open
Recent studies in reinforcement learning (RL) have made significant progress by leveraging function approximation to alleviate the sample complexity hurdle for better performance. Despite the success, existing provably efficient algorithms…
View article: Tractable MCMC for Private Learning with Pure and Gaussian Differential Privacy
Tractable MCMC for Private Learning with Pure and Gaussian Differential Privacy Open
Posterior sampling, i.e., exponential mechanism to sample from the posterior distribution, provides $\varepsilon$-pure differential privacy (DP) guarantees and does not suffer from potentially unbounded privacy breach introduced by $(\vare…
View article: Discovering Mixtures of Structural Causal Models from Time Series Data
Discovering Mixtures of Structural Causal Models from Time Series Data Open
Discovering causal relationships from time series data is significant in fields such as finance, climate science, and neuroscience. However, contemporary techniques rely on the simplifying assumption that data originates from the same caus…
View article: Design and performance of the field cage for the XENONnT experiment
Design and performance of the field cage for the XENONnT experiment Open
The precision in reconstructing events detected in a dual-phase time projection chamber depends on an homogeneous and well understood electric field within the liquid target. In the XENONnT TPC the field homogeneity is achieved through a d…
View article: Deep Bayesian Active Learning for Accelerating Stochastic Simulation
Deep Bayesian Active Learning for Accelerating Stochastic Simulation Open
Stochastic simulations such as large-scale, spatiotemporal, age-structured epidemic models are computationally expensive at fine-grained resolution. While deep surrogate models can speed up the simulations, doing so for stochastic simulati…
View article: Optimization on Pareto sets: On a theory of multi-objective optimization
Optimization on Pareto sets: On a theory of multi-objective optimization Open
In multi-objective optimization, a single decision vector must balance the trade-offs between many objectives. Solutions achieving an optimal trade-off are said to be Pareto optimal: these are decision vectors for which improving any one o…
View article: Linear Convergence of Black-Box Variational Inference: Should We Stick the Landing?
Linear Convergence of Black-Box Variational Inference: Should We Stick the Landing? Open
We prove that black-box variational inference (BBVI) with control variates, particularly the sticking-the-landing (STL) estimator, converges at a geometric (traditionally called "linear") rate under perfect variational family specification…
View article: Reverse Diffusion Monte Carlo
Reverse Diffusion Monte Carlo Open
We propose a Monte Carlo sampler from the reverse diffusion process. Unlike the practice of diffusion models, where the intermediary updates -- the score functions -- are learned with a neural network, we transform the score matching probl…
View article: Langevin Thompson Sampling with Logarithmic Communication: Bandits and Reinforcement Learning
Langevin Thompson Sampling with Logarithmic Communication: Bandits and Reinforcement Learning Open
Thompson sampling (TS) is widely used in sequential decision making due to its ease of use and appealing empirical performance. However, many existing analytical and empirical results for TS rely on restrictive assumptions on reward distri…
View article: A Central Limit Theorem for Algorithmic Estimator of Saddle Point
A Central Limit Theorem for Algorithmic Estimator of Saddle Point Open
In this work, we study the asymptotic randomness of an algorithmic estimator of the saddle point of a globally convex-concave and locally strongly-convex strongly-concave objective. Specifically, we show that the averaged iterates of a Sto…