Search Results for author: Yixin Wang

Found 82 papers, 35 papers with code

Last Layer Empirical Bayes

no code implementations21 May 2025 Valentin Villecroze, Yixin Wang, Gabriel Loaiza-Ganem

Both approaches produce predictions by computing an expectation of neural network outputs over some distribution on the corresponding weights; this distribution is given by the posterior in the case of BNNs, and by a mixture of point masses for ensembles.

Uncertainty Quantification

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model

1 code implementation17 Apr 2025 Zhiwei Xu, Zhiyu Ni, Yixin Wang, Wei Hu

To this end, this paper proposes GrokTransfer, a simple and principled method for accelerating grokking in training neural networks, based on the key observation that data embedding plays a crucial role in determining whether generalization is delayed.

Streamlining Biomedical Research with Specialized LLMs

no code implementations15 Apr 2025 Linqing Chen, Weilei Wang, Yubin Xia, Wentao Wu, Peng Xu, Zilong Bai, Jie Fang, Chaobo Xu, Ran Hu, Licong Xu, Haoran Hua, Jing Sun, Hanmeng Zhong, Jin Liu, Tian Qiu, Haowen Liu, Meng Hu, Xiuwen Li, Fei Gao, Yong Gu, Tao Shi, Chaochao Wang, Jianping Lu, Cheng Sun, Yixin Wang, Shengjie Yang, Yuancheng LI, Lu Jin, Lisha Zhang, Fu Bian, Zhongkai Ye, Lidong Pei, Changyang Tu

In this paper, we propose a novel system that integrates state-of-the-art, domain-specific large language models with advanced information retrieval techniques to deliver comprehensive and context-aware responses.

Decision Making Dialogue Generation +3

Deep Generative Models: Complexity, Dimensionality, and Approximation

1 code implementation1 Apr 2025 Kevin Wang, Hongqian Niu, Yixin Wang, Didong Li

Under this manifold hypothesis, it is widely believed that to approximate a distribution on a $d$-dimensional Riemannian manifold, the latent dimension needs to be at least $d$ or $d+1$.

Doubly robust identification of treatment effects from multiple environments

1 code implementation18 Mar 2025 Piersilvio De Bartolomeis, Julia Kostin, Javier Abad, Yixin Wang, Fanny Yang

Practical and ethical constraints often require the use of observational data for causal inference, particularly in medicine and social sciences.

Causal Inference

Policy Learning with a Natural Language Action Space: A Causal Approach

no code implementations24 Feb 2025 Bohan Zhang, Yixin Wang, Paramveer S. Dhillon

This paper introduces a novel causal framework for multi-stage decision-making in natural language action spaces where outcomes are only observed after a sequence of actions.

Decision Making Q-Learning

Deep Ensembles Secretly Perform Empirical Bayes

no code implementations29 Jan 2025 Gabriel Loaiza-Ganem, Valentin Villecroze, Yixin Wang

The two predominant paradigms to tackle this task are Bayesian neural networks (BNNs) and deep ensembles.

Exponential Family Attention

1 code implementation28 Jan 2025 Kevin Christian Wibisono, Yixin Wang

The self-attention mechanism is the backbone of the transformer neural network underlying most large language models.

Posterior Mean Matching: Generative Modeling through Online Bayesian Inference

no code implementations17 Dec 2024 Sebastian Salazar, Michal Kucer, Yixin Wang, Emily Casleton, David Blei

We demonstrate this flexibility by developing specialized examples: a generative PMM model of real-valued data using the Normal-Normal model, a generative PMM model of count data using a Gamma-Poisson model, and a generative PMM model of discrete data using a Dirichlet-Categorical model.

Bayesian Inference Image Generation +2

Adaptive Nonparametric Perturbations of Parametric Bayesian Models

1 code implementation14 Dec 2024 Bohan Wu, Eli N. Weinstein, Sohrab Salehi, Yixin Wang, David M. Blei

In this paper we study nonparametrically perturbed parametric (NPP) Bayesian models, in which a parametric Bayesian model is relaxed via a distortion of its likelihood.

Bayesian Inference Causal Inference

Ordinal Preference Optimization: Aligning Human Preferences via NDCG

1 code implementation6 Oct 2024 Yang Zhao, Yixin Wang, Mingzhang Yin

In this work, we propose a novel listwise approach named Ordinal Preference Optimization (OPO), which employs the Normalized Discounted Cumulative Gain (NDCG), a widely-used ranking metric, to better utilize relative proximity within ordinal multiple responses.

Information Retrieval

Brain-Cognition Fingerprinting via Graph-GCCA with Contrastive Learning

no code implementations20 Sep 2024 Yixin Wang, Wei Peng, Yu Zhang, Ehsan Adeli, Qingyu Zhao, Kilian M. Pohl

Many longitudinal neuroimaging studies aim to improve the understanding of brain aging and diseases by studying the dynamic interactions between brain function and cognition.

Contrastive Learning Graph Attention

MMedAgent: Learning to Use Medical Tools with Multi-modal Agent

1 code implementation2 Jul 2024 Binxu Li, Tiankai Yan, Yuanting Pan, Jie Luo, Ruiyang Ji, Jiayuan Ding, Zhe Xu, Shilong Liu, Haoyu Dong, Zihao Lin, Yixin Wang

We curate an instruction-tuning dataset comprising six medical tools solving seven tasks across five modalities, enabling the agent to choose the most suitable tools for a given task.

Accuracy on the wrong line: On the pitfalls of noisy data for out-of-distribution generalisation

no code implementations27 Jun 2024 Amartya Sanyal, Yaxi Hu, Yaodong Yu, Yian Ma, Yixin Wang, Bernhard Schölkopf

"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 configurations.

Identifying Nonstationary Causal Structures with High-Order Markov Switching Models

1 code implementation25 Jun 2024 Carles Balsells-Rodas, Yixin Wang, Pedro A. M. Mediano, Yingzhen Li

Causal discovery in time series is a rapidly evolving field with a wide variety of applications in other areas such as climate science and neuroscience.

Causal Discovery Time Series

LLMs Are Prone to Fallacies in Causal Inference

no code implementations18 Jun 2024 Nitish Joshi, Abulhair Saparov, Yixin Wang, He He

Recent work shows that causal facts can be effectively extracted from LLMs through prompting, facilitating the creation of causal graphs for causal inference tasks.

Causal Inference counterfactual

From Unstructured Data to In-Context Learning: Exploring What Tasks Can Be Learned and When

no code implementations31 May 2024 Kevin Christian Wibisono, Yixin Wang

Large language models (LLMs) like transformers demonstrate impressive in-context learning (ICL) capabilities, allowing them to make predictions for new tasks based on prompt exemplars without parameter updates.

In-Context Learning

Addressing Discretization-Induced Bias in Demographic Prediction

1 code implementation27 May 2024 Evan Dong, Aaron Schein, Yixin Wang, Nikhil Garg

In particular, we show that argmax labeling, as used by a prominent commercial voter file vendor to impute race/ethnicity, results in a substantial under-count of African-American voters, e. g., by 28. 2% points in North Carolina.

Imputation Prediction

PatentGPT: A Large Language Model for Intellectual Property

no code implementations28 Apr 2024 Zilong Bai, ruiji zhang, Linqing Chen, Qijun Cai, Yuan Zhong, Cong Wang, Yan Fang, Jie Fang, Jing Sun, Weikuan Wang, Lizhi Zhou, Haoran Hua, Tian Qiu, Chaochao Wang, Cheng Sun, Jianping Lu, Yixin Wang, Yubin Xia, Meng Hu, Haowen Liu, Peng Xu, Licong Xu, Fu Bian, Xiaolong Gu, Lisha Zhang, Weilei Wang, Changyang Tu

In recent years, large language models(LLMs) have attracted significant attention due to their exceptional performance across a multitude of natural language process tasks, and have been widely applied in various fields.

Language Modeling Language Modelling +2

Causal Inference for Human-Language Model Collaboration

1 code implementation30 Mar 2024 Bohan Zhang, Yixin Wang, Paramveer S. Dhillon

A key challenge in answering this causal inference question is formulating an appropriate causal estimand: the conventional average treatment effect (ATE) estimand is inapplicable to text-based treatments due to their high dimensionality.

Causal Inference counterfactual +3

Multi-Domain Causal Representation Learning via Weak Distributional Invariances

no code implementations4 Oct 2023 Kartik Ahuja, Amin Mansouri, Yixin Wang

Causal representation learning has emerged as the center of action in causal machine learning research.

Representation Learning

Rethinking Medical Report Generation: Disease Revealing Enhancement with Knowledge Graph

1 code implementation24 Jul 2023 Yixin Wang, Zihao Lin, Haoyu Dong

Knowledge Graph (KG) plays a crucial role in Medical Report Generation (MRG) because it reveals the relations among diseases and thus can be utilized to guide the generation process.

Medical Report Generation

Bidirectional Attention as a Mixture of Continuous Word Experts

1 code implementation8 Jul 2023 Kevin Christian Wibisono, Yixin Wang

The key observation is that fitting a single-layer single-head bidirectional attention, upon reparameterization, is equivalent to fitting a continuous bag of words (CBOW) model with mixture-of-experts (MoE) weights.

Language Modelling Mixture-of-Experts +2

On the Identifiability of Switching Dynamical Systems

1 code implementation25 May 2023 Carles Balsells-Rodas, Yixin Wang, Yingzhen Li

The identifiability of latent variable models has received increasing attention due to its relevance in interpretability and out-of-distribution generalisation.

Causal Discovery Time Series

Delayed and Indirect Impacts of Link Recommendations

no code implementations17 Mar 2023 Han Zhang, Shangen Lu, Yixin Wang, Mihaela Curmei

Moreover, we show that the effects of recommendations can persist in networks, in part due to their indirect impacts on natural dynamics even after recommendations are turned off.

counterfactual

Team Resilience under Shock: An Empirical Analysis of GitHub Repositories during Early COVID-19 Pandemic

no code implementations29 Jan 2023 Xuan Lu, Wei Ai, Yixin Wang, Qiaozhu Mei

While many organizations have shifted to working remotely during the COVID-19 pandemic, how the remote workforce and the remote teams are influenced by and would respond to this and future shocks remain largely unknown.

counterfactual

Posterior Collapse and Latent Variable Non-identifiability

no code implementations NeurIPS 2021 Yixin Wang, David M. Blei, John P. Cunningham

Unfortunately, variational autoencoders often suffer from posterior collapse: the posterior of the latent variables is equal to its prior, rendering the variational autoencoder useless as a means to produce meaningful representations.

Attribute Variational Inference

Offline Policy Evaluation and Optimization under Confounding

no code implementations29 Nov 2022 Chinmaya Kausik, Yangyi Lu, Kevin Tan, Maggie Makar, Yixin Wang, Ambuj Tewari

Evaluating and optimizing policies in the presence of unobserved confounders is a problem of growing interest in offline reinforcement learning.

Offline RL Off-policy evaluation

Causal Fairness Assessment of Treatment Allocation with Electronic Health Records

no code implementations21 Nov 2022 Linying Zhang, Lauren R. Richter, Yixin Wang, Anna Ostropolets, Noemie Elhadad, David M. Blei, George Hripcsak

Healthcare continues to grapple with the persistent issue of treatment disparities, sparking concerns regarding the equitable allocation of treatments in clinical practice.

Causal Inference Decision Making +1

The Sample Complexity of Online Contract Design

no code implementations10 Nov 2022 Banghua Zhu, Stephen Bates, Zhuoran Yang, Yixin Wang, Jiantao Jiao, Michael I. Jordan

This result shows that exponential-in-$m$ samples are sufficient and necessary to learn a near-optimal contract, resolving an open problem on the hardness of online contract design.

SAP-DETR: Bridging the Gap Between Salient Points and Queries-Based Transformer Detector for Fast Model Convergency

1 code implementation CVPR 2023 Yang Liu, Yao Zhang, Yixin Wang, Yang Zhang, Jiang Tian, Zhongchao shi, Jianping Fan, Zhiqiang He

To bridge the gap between the reference points of salient queries and Transformer detectors, we propose SAlient Point-based DETR (SAP-DETR) by treating object detection as a transformation from salient points to instance objects.

Object object-detection +1

Deep Learning in Single-Cell Analysis

6 code implementations22 Oct 2022 Dylan Molho, Jiayuan Ding, Zhaoheng Li, Hongzhi Wen, Wenzhuo Tang, Yixin Wang, Julian Venegas, Wei Jin, Renming Liu, Runze Su, Patrick Danaher, Robert Yang, Yu Leo Lei, Yuying Xie, Jiliang Tang

Under each task, we describe the most recent developments in classical and deep learning methods and discuss their advantages and disadvantages.

Cell Segmentation Deep Learning +2

Data-Driven Influence Functions for Optimization-Based Causal Inference

no code implementations29 Aug 2022 Michael I. Jordan, Yixin Wang, Angela Zhou

We study a constructive algorithm that approximates Gateaux derivatives for statistical functionals by finite differencing, with a focus on functionals that arise in causal inference.

Causal Inference

Anticipating Performativity by Predicting from Predictions

no code implementations15 Aug 2022 Celestine Mendler-Dünner, Frances Ding, Yixin Wang

Predictions about people, such as their expected educational achievement or their credit risk, can be performative and shape the outcome that they aim to predict.

Valid Inference After Causal Discovery

no code implementations11 Aug 2022 Paula Gradu, Tijana Zrnic, Yixin Wang, Michael I. Jordan

Causal discovery and causal effect estimation are two fundamental tasks in causal inference.

Causal Discovery Causal Inference +1

Statistical and Computational Trade-offs in Variational Inference: A Case Study in Inferential Model Selection

no code implementations22 Jul 2022 Kush Bhatia, Nikki Lijing Kuang, Yi-An Ma, Yixin Wang

Focusing on Gaussian inferential models (or variational approximating families) with diagonal plus low-rank precision matrices, we initiate a theoretical study of the trade-offs in two aspects, Bayesian posterior inference error and frequentist uncertainty quantification error.

Bayesian Inference Computational Efficiency +4

Recommendation Systems with Distribution-Free Reliability Guarantees

no code implementations4 Jul 2022 Anastasios N. Angelopoulos, Karl Krauth, Stephen Bates, Yixin Wang, Michael I. Jordan

Building from a pre-trained ranking model, we show how to return a set of items that is rigorously guaranteed to contain mostly good items.

Diversity Learning-To-Rank +1

Breaking Feedback Loops in Recommender Systems with Causal Inference

no code implementations4 Jul 2022 Karl Krauth, Yixin Wang, Michael I. Jordan

Our main observation is that a recommender system does not suffer from feedback loops if it reasons about causal quantities, namely the intervention distributions of recommendations on user ratings.

Causal Inference Recommendation Systems

Partial Identification with Noisy Covariates: A Robust Optimization Approach

no code implementations22 Feb 2022 Wenshuo Guo, Mingzhang Yin, Yixin Wang, Michael I. Jordan

Directly adjusting for these imperfect measurements of the covariates can lead to biased causal estimates.

Causal Inference

Augmenting Neural Networks with Priors on Function Values

no code implementations10 Feb 2022 Hunter Nisonoff, Yixin Wang, Jennifer Listgarten

The need for function estimation in label-limited settings is common in the natural sciences.

Spatiotemporal Clustering with Neyman-Scott Processes via Connections to Bayesian Nonparametric Mixture Models

1 code implementation13 Jan 2022 Yixin Wang, Anthony Degleris, Alex H. Williams, Scott W. Linderman

This construction is similar to Bayesian nonparametric mixture models like the Dirichlet process mixture model (DPMM) in that the number of latent events (i. e. clusters) is a random variable, but the point process formulation makes the NSP especially well suited to modeling spatiotemporal data.

Bayesian Inference Clustering +1

A Survey of Visual Transformers

1 code implementation11 Nov 2021 Yang Liu, Yao Zhang, Yixin Wang, Feng Hou, Jin Yuan, Jiang Tian, Yang Zhang, Zhongchao shi, Jianping Fan, Zhiqiang He

Transformer, an attention-based encoder-decoder model, has already revolutionized the field of natural language processing (NLP).

Articles Survey

Identifiable Deep Generative Models via Sparse Decoding

1 code implementation20 Oct 2021 Gemma E. Moran, Dhanya Sridhar, Yixin Wang, David M. Blei

The underlying model is sparse in that each observed feature (i. e. each dimension of the data) depends on a small subset of the latent factors.

Representation Learning

All-Around Real Label Supervision: Cyclic Prototype Consistency Learning for Semi-supervised Medical Image Segmentation

1 code implementation28 Sep 2021 Zhe Xu, Yixin Wang, Donghuan Lu, Lequan Yu, Jiangpeng Yan, Jie Luo, Kai Ma, Yefeng Zheng, Raymond Kai-yu Tong

Observing this, we ask an unexplored but interesting question: can we exploit the unlabeled data via explicit real label supervision for semi-supervised training?

All Brain Tumor Segmentation +4

Desiderata for Representation Learning: A Causal Perspective

1 code implementation8 Sep 2021 Yixin Wang, Michael I. Jordan

Representation learning constructs low-dimensional representations to summarize essential features of high-dimensional data.

counterfactual Disentanglement

Learning Equilibria in Matching Markets from Bandit Feedback

no code implementations NeurIPS 2021 Meena Jagadeesan, Alexander Wei, Yixin Wang, Michael I. Jordan, Jacob Steinhardt

Large-scale, two-sided matching platforms must find market outcomes that align with user preferences while simultaneously learning these preferences from data.

ACN: Adversarial Co-training Network for Brain Tumor Segmentation with Missing Modalities

2 code implementations28 Jun 2021 Yixin Wang, Yang Zhang, Yang Liu, Zihao Lin, Jiang Tian, Cheng Zhong, Zhongchao shi, Jianping Fan, Zhiqiang He

Specifically, ACN adopts a novel co-training network, which enables a coupled learning process for both full modality and missing modality to supplement each other's domain and feature representations, and more importantly, to recover the `missing' information of absent modalities.

Brain Tumor Segmentation Transfer Learning +1

Trust It or Not: Confidence-Guided Automatic Radiology Report Generation

no code implementations21 Jun 2021 Yixin Wang, Zihao Lin, Zhe Xu, Haoyu Dong, Jiang Tian, Jie Luo, Zhongchao shi, Yang Zhang, Jianping Fan, Zhiqiang He

Experimental results have demonstrated that the proposed method for model uncertainty characterization and estimation can produce more reliable confidence scores for radiology report generation, and the modified loss function, which takes into account the uncertainties, leads to better model performance on two public radiology report datasets.

Decision Making Image Captioning +2

Noisy Labels are Treasure: Mean-Teacher-Assisted Confident Learning for Hepatic Vessel Segmentation

1 code implementation3 Jun 2021 Zhe Xu, Donghuan Lu, Yixin Wang, Jie Luo, Jayender Jagadeesan, Kai Ma, Yefeng Zheng, Xiu Li

Manually segmenting the hepatic vessels from Computer Tomography (CT) is far more expertise-demanding and laborious than other structures due to the low-contrast and complex morphology of vessels, resulting in the extreme lack of high-quality labeled data.

Multi-Source Causal Inference Using Control Variates

no code implementations30 Mar 2021 Wenshuo Guo, Serena Wang, Peng Ding, Yixin Wang, Michael I. Jordan

Across simulations and two case studies with real data, we show that this control variate can significantly reduce the variance of the ATE estimate.

Causal Inference Epidemiology +2

Semi-supervised Cardiac Image Segmentation via Label Propagation and Style Transfer

1 code implementation29 Dec 2020 Yao Zhang, Jiawei Yang, Feng Hou, Yang Liu, Yixin Wang, Jiang Tian, Cheng Zhong, Yang Zhang, Zhiqiang He

Accurate segmentation of cardiac structures can assist doctors to diagnose diseases, and to improve treatment planning, which is highly demanded in the clinical practice.

Image Segmentation Segmentation +2

Double-Uncertainty Weighted Method for Semi-supervised Learning

no code implementations19 Oct 2020 Yixin Wang, Yao Zhang, Jiang Tian, Cheng Zhong, Zhongchao shi, Yang Zhang, Zhiqiang He

We train the teacher model using Bayesian deep learning to obtain double-uncertainty, i. e. segmentation uncertainty and feature uncertainty.

Segmentation

Towards Clarifying the Theory of the Deconfounder

no code implementations10 Mar 2020 Yixin Wang, David M. Blei

Wang and Blei (2019) studies multiple causal inference and proposes the deconfounder algorithm.

Causal Inference

The Blessings of Multiple Causes: A Reply to Ogburn et al. (2019)

no code implementations15 Oct 2019 Yixin Wang, David M. Blei

Ogburn et al. (2019, arXiv:1910. 05438) discuss "The Blessings of Multiple Causes" (Wang and Blei, 2018, arXiv:1805. 06826).

Cascaded Volumetric Convolutional Network for Kidney Tumor Segmentation from CT volumes

no code implementations5 Oct 2019 Yao Zhang, Yixin Wang, Feng Hou, Jiawei Yang, Guangwei Xiong, Jiang Tian, Cheng Zhong

Automated segmentation of kidney and tumor from 3D CT scans is necessary for the diagnosis, monitoring, and treatment planning of the disease.

Segmentation Tumor Segmentation

Multiple Causes: A Causal Graphical View

no code implementations30 May 2019 Yixin Wang, David M. Blei

Our results expand the theory in Wang & Blei (2018), justify the deconfounder for causal graphs, and extend the settings where it can be used.

Causal Inference valid

Equal Opportunity and Affirmative Action via Counterfactual Predictions

no code implementations26 May 2019 Yixin Wang, Dhanya Sridhar, David M. Blei

Machine learning (ML) can automate decision-making by learning to predict decisions from historical data.

counterfactual Decision Making +1

Variational Bayes under Model Misspecification

1 code implementation NeurIPS 2019 Yixin Wang, David M. Blei

As a consequence of these results, we find that the model misspecification error dominates the variational approximation error in VB posterior predictive distributions.

model

The Medical Deconfounder: Assessing Treatment Effects with Electronic Health Records

no code implementations3 Apr 2019 Linying Zhang, Yixin Wang, Anna Ostropolets, Jami J. Mulgrave, David M. Blei, George Hripcsak

To adjust for unobserved confounders, we develop the medical deconfounder, a machine learning algorithm that unbiasedly estimates treatment effects from EHRs.

The Deconfounded Recommender: A Causal Inference Approach to Recommendation

no code implementations20 Aug 2018 Yixin Wang, Dawen Liang, Laurent Charlin, David M. Blei

To this end, we develop a causal approach to recommendation, one where watching a movie is a "treatment" and a user's rating is an "outcome."

Causal Inference Recommendation Systems

Black Box FDR

no code implementations ICML 2018 Wesley Tansey, Yixin Wang, David M. Blei, Raul Rabadan

BB-FDR learns a series of black box predictive models to boost power and control the false discovery rate (FDR) at two stages of study analysis.

The Blessings of Multiple Causes

2 code implementations17 May 2018 Yixin Wang, David M. Blei

Causal inference from observational data often assumes "ignorability," that all confounders are observed.

Causal Inference

Frequentist Consistency of Variational Bayes

no code implementations9 May 2017 Yixin Wang, David M. Blei

The theorem leverages the theoretical characterizations of frequentist variational approximations to understand asymptotic properties of VB.

Minimal Dispersion Approximately Balancing Weights: Asymptotic Properties and Practical Considerations

1 code implementation2 May 2017 Yixin Wang, José R. Zubizarreta

The key observation is the connection between approximate covariate balance and shrinkage estimation of the propensity score.

Methodology Statistics Theory Applications Statistics Theory

Robust Probabilistic Modeling with Bayesian Data Reweighting

1 code implementation ICML 2017 Yixin Wang, Alp Kucukelbir, David M. Blei

We propose a way to systematically detect and mitigate mismatch of a large class of probabilistic models.

Cannot find the paper you are looking for? You can Submit a new open access paper.