no code implementations • ACL (WebNLG, INLG) 2020 • Qipeng Guo, Zhijing Jin, Ning Dai, Xipeng Qiu, xiangyang xue, David Wipf, Zheng Zhang
Text verbalization of knowledge graphs is an important problem with wide application to natural language generation (NLG) systems.
no code implementations • 1 Feb 2025 • Yixuan He, Aaron Sandel, David Wipf, Mihai Cucuringu, John Mitani, Gesine Reinert
How can we identify groups of primate individuals which could be conjectured to drive social structure?
1 code implementation • 13 Oct 2024 • Yanlin Zhang, Ning li, Quan Gan, Weinan Zhang, David Wipf, Minjie Wang
But despite this potential, evaluations thus far are primarily based on the end performance of a complete ML pipeline, providing limited insight into precisely how LLMs behave relative to human experts in feature engineering.
1 code implementation • 13 Sep 2024 • Qitian Wu, Kai Yang, Hengrui Zhang, David Wipf, Junchi Yan
Learning representations on large graphs is a long-standing challenge due to the inter-dependence nature.
no code implementations • 13 Sep 2024 • Qitian Wu, David Wipf, Junchi Yan
Learning representations for structured data with certain geometries (observed or unobserved) is a fundamental challenge, wherein message passing neural networks (MPNNs) have become a de facto class of model solutions.
no code implementations • 12 Jul 2024 • Xiangkun Hu, Tong He, David Wipf
Large language models in the past have typically relied on some form of reinforcement learning with human feedback (RLHF) to better align model responses with human preferences.
1 code implementation • 28 Apr 2024 • Minjie Wang, Quan Gan, David Wipf, Zhenkun Cai, Ning li, Jianheng Tang, Yanlin Zhang, Zizhao Zhang, Zunyao Mao, Yakun Song, Yanbo Wang, Jiahang Li, Han Zhang, Guang Yang, Xiao Qin, Chuan Lei, Muhan Zhang, Weinan Zhang, Christos Faloutsos, Zheng Zhang
Although RDBs store vast amounts of rich, informative data spread across interconnected tables, the progress of predictive machine learning models as applied to such tasks arguably falls well behind advances in other domains such as computer vision or natural language processing.
1 code implementation • 7 Mar 2024 • Amber Yijia Zheng, Tong He, Yixuan Qiu, Minjie Wang, David Wipf
These optimal features typically depend on tunable parameters of the lower-level energy in such a way that the entire bilevel pipeline can be trained end-to-end.
no code implementations • 4 Dec 2023 • Han Zhang, Quan Gan, David Wipf, Weinan Zhang
Consequently, the prevalent approach for training machine learning models on data stored in relational databases involves performing feature engineering to merge the data from multiple tables into a single table and subsequently applying single table models.
no code implementations • 19 Oct 2023 • Haitian Jiang, Renjie Liu, Xiao Yan, Zhenkun Cai, Minjie Wang, David Wipf
Among the many variants of graph neural network (GNN) architectures capable of modeling data with cross-instance relations, an important subclass involves layers designed such that the forward pass iteratively reduces a graph-regularized energy function of interest.
1 code implementation • 14 Oct 2023 • Yuxin Wang, Xiannian Hu, Quan Gan, Xuanjing Huang, Xipeng Qiu, David Wipf
In contrast, \emph{edge-wise} methods rely on the formation of edge-specific subgraph embeddings to enrich the representation of pair-wise relationships, disambiguating isomorphic nodes to improve accuracy, but with increased model complexity.
1 code implementation • 9 Oct 2023 • Yixuan He, Gesine Reinert, David Wipf, Mihai Cucuringu
The angular synchronization problem aims to accurately estimate (up to a constant additive phase) a set of unknown angles $\theta_1, \dots, \theta_n\in[0, 2\pi)$ from $m$ noisy measurements of their offsets $\theta_i-\theta_j \;\mbox{mod} \; 2\pi.$ Applications include, for example, sensor network localization, phase retrieval, and distributed clock synchronization.
1 code implementation • 8 Oct 2023 • Chenxiao Yang, Qitian Wu, David Wipf, Ruoyu Sun, Junchi Yan
In particular, we find that the gradient descent optimization of GNNs implicitly leverages the graph structure to update the learned function, as can be quantified by a phenomenon which we call \emph{kernel-graph alignment}.
1 code implementation • 16 Jun 2023 • Yuxin Wang, Quan Gan, Xipeng Qiu, Xuanjing Huang, David Wipf
Hypergraphs are a powerful abstraction for representing higher-order interactions between entities of interest.
1 code implementation • 14 Jun 2023 • Qitian Wu, Wentao Zhao, Zenan Li, David Wipf, Junchi Yan
In this paper, we introduce a novel all-pair message passing scheme for efficiently propagating node signals between arbitrary nodes, as an important building block for a pioneering Transformer-style network for node classification on large graphs, dubbed as \textsc{NodeFormer}.
1 code implementation • 23 Feb 2023 • Yijia Zheng, Tong He, Yixuan Qiu, David Wipf
Although the variational autoencoder (VAE) and its conditional extension (CVAE) are capable of state-of-the-art results across multiple domains, their precise behavior is still not fully understood, particularly in the context of data (like images) that lie on or near a low-dimensional manifold.
1 code implementation • 23 Jan 2023 • Qitian Wu, Chenxiao Yang, Wentao Zhao, Yixuan He, David Wipf, Junchi Yan
Real-world data generation often involves complex inter-dependencies among instances, violating the IID-data hypothesis of standard learning paradigms and posing a challenge for uncovering the geometric structures for learning desired instance representations.
no code implementations • 18 Jan 2023 • Kezhao Huang, Haitian Jiang, Minjie Wang, Guangxuan Xiao, David Wipf, Xiang Song, Quan Gan, Zengfeng Huang, Jidong Zhai, Zheng Zhang
A key performance bottleneck when training graph neural network (GNN) models on large, real-world graphs is loading node features onto a GPU.
no code implementations • 25 Dec 2022 • Jiarui Jin, Yangkun Wang, Weinan Zhang, Quan Gan, Xiang Song, Yong Yu, Zheng Zhang, David Wipf
However, existing methods lack elaborate design regarding the distinctions between two tasks that have been frequently overlooked: (i) edges only constitute the topology in the node classification task but can be used as both the topology and the supervisions (i. e., labels) in the edge prediction task; (ii) the node classification makes prediction over each individual node, while the edge prediction is determinated by each pair of nodes.
1 code implementation • 23 Oct 2022 • Jian Yao, Yuxin Hong, Chiyu Wang, Tianjun Xiao, Tong He, Francesco Locatello, David Wipf, Yanwei Fu, Zheng Zhang
The key intuition is that the occluded part of an object can be explained away if that part is visible in other frames, possibly deformed as long as the deformation can be reasonably learned.
1 code implementation • 22 Jun 2022 • Hongjoon Ahn, Yongyi Yang, Quan Gan, Taesup Moon, David Wipf
Moreover, the complexity of this trade-off is compounded in the heterogeneous graph case due to the disparate heterophily relationships between nodes of different types.
no code implementations • 16 Jun 2022 • Jiuhai Chen, Jonas Mueller, Vassilis N. Ioannidis, Tom Goldstein, David Wipf
Graph Neural Networks (GNNs) with numerical node features and graph structure as inputs have demonstrated superior performance on various supervised learning tasks with graph data.
1 code implementation • 14 Jun 2022 • Kounianhua Du, Weinan Zhang, Ruiwen Zhou, Yangkun Wang, Xilong Zhao, Jiarui Jin, Quan Gan, Zheng Zhang, David Wipf
Prediction over tabular data is an essential and fundamental problem in many important downstream tasks.
1 code implementation • 27 May 2022 • Yongyi Yang, Zengfeng Huang, David Wipf
Deep learning models such as the Transformer are often constructed by heuristics and experience.
2 code implementations • ICLR 2022 • Qitian Wu, Hengrui Zhang, Junchi Yan, David Wipf
There is increasing evidence suggesting neural networks' sensitivity to distribution shifts, so that research on out-of-distribution (OOD) generalization comes into the spotlight.
1 code implementation • 1 Feb 2022 • Yixuan He, Quan Gan, David Wipf, Gesine Reinert, Junchi Yan, Mihai Cucuringu
In this paper, we introduce neural networks into the ranking recovery problem by proposing the so-called GNNRank, a trainable GNN-based framework with digraph embedding.
no code implementations • NeurIPS 2021 • Longyuan Li, Jian Yao, Li Wenliang, Tong He, Tianjun Xiao, Junchi Yan, David Wipf, Zheng Zhang
Learning the distribution of future trajectories conditioned on the past is a crucial problem for understanding multi-agent systems.
no code implementations • NeurIPS 2021 • Bin Dai, Li Wenliang, David Wipf
A number of recent studies of continuous variational autoencoder (VAE) models have noted, either directly or indirectly, the tendency of various parameter gradients to drift towards infinity during training.
no code implementations • 12 Nov 2021 • Yongyi Yang, Tang Liu, Yangkun Wang, Zengfeng Huang, David Wipf
It has been observed that graph neural networks (GNN) sometimes struggle to maintain a healthy balance between the efficient modeling long-range dependencies across nodes while avoiding unintended consequences such oversmoothed node representations or sensitivity to spurious edges.
1 code implementation • 26 Oct 2021 • Jiuhai Chen, Jonas Mueller, Vassilis N. Ioannidis, Soji Adeshina, Yangkun Wang, Tom Goldstein, David Wipf
For supervised learning with tabular data, decision tree ensembles produced via boosting techniques generally dominate real-world applications involving iid training/test sets.
1 code implementation • ICLR 2022 • Yangkun Wang, Jiarui Jin, Weinan Zhang, Yongyi Yang, Jiuhai Chen, Quan Gan, Yong Yu, Zheng Zhang, Zengfeng Huang, David Wipf
In this regard, it has recently been proposed to use a randomly-selected portion of the training labels as GNN inputs, concatenated with the original node features for making predictions on the remaining labels.
no code implementations • 29 Sep 2021 • Mucong Ding, Kezhi Kong, Jiuhai Chen, John Kirchenbauer, Micah Goldblum, David Wipf, Furong Huang, Tom Goldstein
We observe that in most cases, we need both a suitable domain generalization algorithm and a strong GNN backbone model to optimize out-of-distribution test performance.
no code implementations • 29 Sep 2021 • Hengrui Zhang, Qitian Wu, Shaofeng Zhang, Junchi Yan, David Wipf, Philip S. Yu
In this paper, we propose ESCo (Effective and Scalable Contrastive), a new contrastive framework which is essentially an instantiation of the Information Bottleneck principle under self-supervised learning settings.
no code implementations • ICLR 2022 • Jiarui Jin, Yangkun Wang, Kounianhua Du, Weinan Zhang, Zheng Zhang, David Wipf, Yong Yu, Quan Gan
Prevailing methods for relation prediction in heterogeneous graphs aim at learning latent representations (i. e., embeddings) of observed nodes and relations, and thus are limited to the transductive setting where the relation types must be known during training.
no code implementations • ICLR 2022 • Jiuhai Chen, Jonas Mueller, Vassilis N. Ioannidis, Soji Adeshina, Yangkun Wang, Tom Goldstein, David Wipf
Many practical modeling tasks require making predictions using tabular data composed of heterogeneous feature types (e. g., text-based, categorical, continuous, etc.).
3 code implementations • ICCV 2021 • Yifan Xing, Tong He, Tianjun Xiao, Yongxin Wang, Yuanjun Xiong, Wei Xia, David Wipf, Zheng Zhang, Stefano Soatto
Our hierarchical GNN uses a novel approach to merge connected components predicted at each level of the hierarchy to form a new graph at the next level.
1 code implementation • NeurIPS 2021 • Hengrui Zhang, Qitian Wu, Junchi Yan, David Wipf, Philip S. Yu
We introduce a conceptually simple yet effective model for self-supervised representation learning with graph data.
no code implementations • CVPR 2021 • Yu Zhang, Daniel Lau, David Wipf
Three-dimensional scanning by means of structured light illumination is an active imaging technique involving projecting and capturing a series of striped patterns and then using the observed warping of stripes to reconstruct the target object's surface through triangulating each pixel in the camera to a unique projector coordinate corresponding to a particular feature in the projected patterns.
2 code implementations • 24 Mar 2021 • Yangkun Wang, Jiarui Jin, Weinan Zhang, Yong Yu, Zheng Zhang, David Wipf
Over the past few years, graph neural networks (GNN) and label propagation-based methods have made significant progress in addressing node classification tasks on graphs.
Ranked #1 on
Node Property Prediction
on ogbn-proteins
1 code implementation • 10 Mar 2021 • Yongyi Yang, Tang Liu, Yangkun Wang, Jinjing Zhou, Quan Gan, Zhewei Wei, Zheng Zhang, Zengfeng Huang, David Wipf
Despite the recent success of graph neural networks (GNN), common architectures often exhibit significant limitations, including sensitivity to oversmoothing, long-range dependencies, and spurious edges, e. g., as can occur as a result of graph heterophily or adversarial attacks.
1 code implementation • NeurIPS 2021 • Qingru Zhang, David Wipf, Quan Gan, Le Song
Graph neural networks (GNN) have recently emerged as a vehicle for applying deep network architectures to graph and relational data.
no code implementations • 1 Jan 2021 • Jiarui Jin, Cong Chen, Ming Zhou, Weinan Zhang, Rasool Fakoor, David Wipf, Yong Yu, Jun Wang, Alex Smola
Goal-oriented reinforcement learning algorithms are often good at exploration, not exploitation, while episodic algorithms excel at exploitation, not exploration.
no code implementations • 1 Jan 2021 • Jiarui Jin, Sijin Zhou, Weinan Zhang, Rasool Fakoor, David Wipf, Tong He, Yong Yu, Zheng Zhang, Alex Smola
In reinforcement learning, a map with states and transitions built based on historical trajectories is often helpful in exploration and exploitation.
1 code implementation • 14 Dec 2020 • Qipeng Guo, Zhijing Jin, Ziyu Wang, Xipeng Qiu, Weinan Zhang, Jun Zhu, Zheng Zhang, David Wipf
Cycle-consistent training is widely used for jointly learning a forward and inverse mapping between two domains of interest without the cumbersome requirement of collecting matched pairs within each domain.
1 code implementation • NeurIPS 2020 • Ziyu Wang, Bin Dai, David Wipf, Jun Zhu
The recent, counter-intuitive discovery that deep generative models (DGMs) can frequently assign a higher likelihood to outliers has implications for both outlier detection applications as well as our overall understanding of generative modeling.
2 code implementations • ACL (WebNLG, INLG) 2020 • Qipeng Guo, Zhijing Jin, Xipeng Qiu, Wei-Nan Zhang, David Wipf, Zheng Zhang
Due to the difficulty and high cost of data collection, the supervised data available in the two fields are usually on the magnitude of tens of thousands, for example, 18K in the WebNLG~2017 dataset after preprocessing, which is far fewer than the millions of data for other tasks such as machine translation.
no code implementations • ICML 2020 • Bin Dai, Ziyu Wang, David Wipf
In narrow asymptotic settings Gaussian VAE models of continuous data have been shown to possess global optima aligned with ground-truth distributions.
1 code implementation • CVPR 2019 • Kaixuan Wei, Jiaolong Yang, Ying Fu, David Wipf, Hua Huang
Removing undesirable reflections from a single image captured through a glass window is of practical importance to visual computing systems.
Ranked #3 on
Reflection Removal
on SIR^2(Objects)
4 code implementations • ICLR 2019 • Bin Dai, David Wipf
Although variational autoencoders (VAEs) represent a widely influential deep generative model, many aspects of the underlying energy function remain poorly understood.
1 code implementation • 7 Nov 2018 • Qingnan Fan, Jiaolong Yang, David Wipf, Baoquan Chen, Xin Tong
Image smoothing represents a fundamental component of many disparate computer vision and graphics applications.
1 code implementation • ICML 2018 • Bin Dai, Chen Zhu, Baining Guo, David Wipf
Neural networks can be compressed to reduce memory and computational requirements, or to increase accuracy by facilitating the use of a larger base architecture.
1 code implementation • ICML 2018 • Bin Dai, Chen Zhu, David Wipf
Neural networks can be compressed to reduce memory and computational requirements, or to increase accuracy by facilitating the use of a larger base architecture.
1 code implementation • ICCV 2017 • Qingnan Fan, Jiaolong Yang, Gang Hua, Baoquan Chen, David Wipf
This paper proposes a deep neural network structure that exploits edge information in addressing representative low-level vision tasks such as layer separation and image filtering.
1 code implementation • 16 Jun 2017 • Bin Dai, Yu Wang, John Aston, Gang Hua, David Wipf
Variational autoencoders (VAE) represent a popular, flexible form of deep generative model that can be stochastically fit to samples from a given random process using an information-theoretic variational bound on the true underlying distribution.
no code implementations • NeurIPS 2017 • Hao He, Bo Xin, David Wipf
The iterations of many first-order algorithms, when applied to minimizing common regularized regression functions, often resemble neural network layers with pre-specified weights.
no code implementations • CVPR 2018 • Qingnan Fan, Jiaolong Yang, Gang Hua, Baoquan Chen, David Wipf
While invaluable for many computer vision applications, decomposing a natural image into intrinsic reflectance and shading layers represents a challenging, underdetermined inverse problem.
no code implementations • NeurIPS 2016 • Tae-Hyun Oh, Yasuyuki Matsushita, In Kweon, David Wipf
Commonly used in many applications, robust PCA represents an algorithmic attempt to reduce the sensitivity of classical PCA to outliers.
no code implementations • NeurIPS 2016 • Bo Xin, Yizhou Wang, Wen Gao, David Wipf
The iterations of many sparse estimation algorithms are comprised of a fixed linear filter cascaded with a thresholding nonlinearity, which collectively resemble a typical neural network layer.
no code implementations • 7 Dec 2015 • Tae-Hyun Oh, Yasuyuki Matsushita, In So Kweon, David Wipf
Commonly used in computer vision and other applications, robust PCA represents an algorithmic attempt to reduce the sensitivity of classical PCA to outliers.
no code implementations • ICCV 2015 • Huan Yang, Baoyuan Wang, Stephen Lin, David Wipf, Minyi Guo, Baining Guo
With the growing popularity of short-form video sharing platforms such as \em{Instagram} and \em{Vine}, there has been an increasing need for techniques that automatically extract highlights from video.
no code implementations • 9 Aug 2014 • David Wipf
In many applications that require matrix solutions of minimal rank, the underlying cost function is non-convex leading to an intractable, NP-hard optimization problem.
no code implementations • 10 Jun 2014 • Bo Xin, David Wipf
While elegant theoretical conditions elucidate when this replacement is likely to be successful, they are highly restrictive and convex algorithms fail when the ambient rank is too high or when the constraint set is poorly structured.
no code implementations • NeurIPS 2013 • Haichao Zhang, David Wipf
Typical blur from camera shake often deviates from the standard uniform convolutional assumption, in part because of problematic rotations which create greater blurring away from some unknown center point.
no code implementations • 17 Jun 2013 • Haichao Zhang, David Wipf
Typical blur from camera shake often deviates from the standard uniform convolutional script, in part because of problematic rotations which create greater blurring away from some unknown center point.
no code implementations • CVPR 2013 • Haichao Zhang, David Wipf, Yanning Zhang
This paper presents a robust algorithm for estimating a single latent sharp image given multiple blurry and/or noisy observations.
no code implementations • 10 May 2013 • David Wipf, Haichao Zhang
Blind deconvolution involves the estimation of a sharp signal or image given only a blurry observation.