Search Results for author: Danai Koutra

Found 48 papers, 26 papers with code

TouchUp-G: Improving Feature Representation through Graph-Centric Finetuning

no code implementations25 Sep 2023 Jing Zhu, Xiang Song, Vassilis N. Ioannidis, Danai Koutra, Christos Faloutsos

How can we enhance the node features acquired from Pretrained Models (PMs) to better suit downstream graph learning tasks?

Domain Adaptation Graph Learning +2

Accurate and Scalable Estimation of Epistemic Uncertainty for Graph Neural Networks

no code implementations20 Sep 2023 Puja Trivedi, Mark Heimann, Rushil Anirudh, Danai Koutra, Jayaraman J. Thiagarajan

Safe deployment of graph neural networks (GNNs) under distribution shift requires models to provide accurate confidence indicators (CI).

Interpretable Sparsification of Brain Graphs: Better Practices and Effective Designs for Graph Neural Networks

1 code implementation26 Jun 2023 Gaotang Li, Marlena Duda, Xiang Zhang, Danai Koutra, Yujun Yan

Based on these insights, we propose a new model, Interpretable Graph Sparsification (IGS), which enhances graph classification performance by up to 5. 1% with 55. 0% fewer edges.

Graph Classification

On Performance Discrepancies Across Local Homophily Levels in Graph Neural Networks

no code implementations8 Jun 2023 Donald Loveland, Jiong Zhu, Mark Heimann, Benjamin Fish, Michael T. Shaub, Danai Koutra

We ground the practical implications of this work through granular analysis on five real-world datasets with varying global homophily levels, demonstrating that (a) GNNs can fail to generalize to test nodes that deviate from the global homophily of a graph, and (b) high local homophily does not necessarily confer high performance for a node.

Node Classification

SpotTarget: Rethinking the Effect of Target Edges for Link Prediction in Graph Neural Networks

no code implementations1 Jun 2023 Jing Zhu, YuHang Zhou, Vassilis N. Ioannidis, Shengyi Qian, Wei Ai, Xiang Song, Danai Koutra

Particularly, in prevalent GNN frameworks (e. g., DGL and PyTorch-Geometric), the target edges (i. e., the edges being predicted) consistently exist as message passing edges in the graph during training.

Link Prediction Node Classification

Size Generalizability of Graph Neural Networks on Biological Data: Insights and Practices from the Spectral Perspective

no code implementations24 May 2023 Yujun Yan, Gaotang Li, Danai Koutra

We identify a distribution shift between small and large graphs in the eigenvalues of the normalized Laplacian/adjacency matrix, indicating a difference in the global node connectivity, which is found to be correlated with the node closeness centrality.

Graph Classification

Simplifying Distributed Neural Network Training on Massive Graphs: Randomized Partitions Improve Model Aggregation

1 code implementation17 May 2023 Jiong Zhu, Aishwarya Reganti, Edward Huang, Charles Dickens, Nikhil Rao, Karthik Subbian, Danai Koutra

Backed by our theoretical analysis, instead of maximizing the recovery of cross-instance node dependencies -- which has been considered the key behind closing the performance gap between model aggregation and centralized training -- , our framework leverages randomized assignment of nodes or super-nodes (i. e., collections of original nodes) to partition the training graph such that it improves data uniformity and minimizes the discrepancy of gradient and loss function across instances.

A Closer Look at Model Adaptation using Feature Distortion and Simplicity Bias

no code implementations23 Mar 2023 Puja Trivedi, Danai Koutra, Jayaraman J. Thiagarajan

Advances in the expressivity of pretrained models have increased interest in the design of adaptation protocols which enable safe and effective transfer learning.

Out-of-Distribution Generalization Transfer Learning

On the Efficacy of Generalization Error Prediction Scoring Functions

no code implementations23 Mar 2023 Puja Trivedi, Danai Koutra, Jayaraman J. Thiagarajan

Overall, our work carefully studies the effectiveness of popular scoring functions in realistic settings and helps to better understand their limitations.

CAPER: Coarsen, Align, Project, Refine - A General Multilevel Framework for Network Alignment

1 code implementation23 Aug 2022 Jing Zhu, Danai Koutra, Mark Heimann

Network alignment, or the task of finding corresponding nodes in different networks, is an important problem formulation in many application domains.

Analyzing Data-Centric Properties for Graph Contrastive Learning

1 code implementation4 Aug 2022 Puja Trivedi, Ekdeep Singh Lubana, Mark Heimann, Danai Koutra, Jayaraman J. Thiagarajan

Overall, our work rigorously contextualizes, both empirically and theoretically, the effects of data-centric properties on augmentation strategies and learning paradigms for graph SSL.

Contrastive Learning Self-Supervised Learning +1

Exploring the Design of Adaptation Protocols for Improved Generalization and Machine Learning Safety

no code implementations26 Jul 2022 Puja Trivedi, Danai Koutra, Jayaraman J. Thiagarajan

While directly fine-tuning (FT) large-scale, pretrained models on task-specific data is well-known to induce strong in-distribution task performance, recent works have demonstrated that different adaptation protocols, such as linear probing (LP) prior to FT, can improve out-of-distribution generalization.

Anomaly Detection BIG-bench Machine Learning +2

On Graph Neural Network Fairness in the Presence of Heterophilous Neighborhoods

no code implementations10 Jul 2022 Donald Loveland, Jiong Zhu, Mark Heimann, Ben Fish, Michael T. Schaub, Danai Koutra

We study the task of node classification for graph neural networks (GNNs) and establish a connection between group fairness, as measured by statistical parity and equal opportunity, and local assortativity, i. e., the tendency of linked nodes to have similar attributes.

Fairness Node Classification

Learning node embeddings via summary graphs: a brief theoretical analysis

no code implementations4 Jul 2022 Houquan Zhou, Shenghua Liu, Danai Koutra, HuaWei Shen, Xueqi Cheng

Recent works try to improve scalability via graph summarization -- i. e., they learn embeddings on a smaller summary graph, and then restore the node embeddings of the original graph.

Graph Mining Graph Representation Learning

Leveraging the Graph Structure of Neural Network Training Dynamics

1 code implementation9 Nov 2021 Fatemeh Vahedian, Ruiyu Li, Puja Trivedi, Di Jin, Danai Koutra

Understanding the training dynamics of deep neural networks (DNNs) is important as it can lead to improved training efficiency and task performance.

Augmentations in Graph Contrastive Learning: Current Methodological Flaws & Towards Better Practices

no code implementations5 Nov 2021 Puja Trivedi, Ekdeep Singh Lubana, Yujun Yan, Yaoqing Yang, Danai Koutra

Unsupervised graph representation learning is critical to a wide range of applications where labels may be scarce or expensive to procure.

Contrastive Learning Data Augmentation +5

Deep Transfer Learning for Multi-source Entity Linkage via Domain Adaptation

1 code implementation27 Oct 2021 Di Jin, Bunyamin Sisman, Hao Wei, Xin Luna Dong, Danai Koutra

AdaMEL models the attribute importance that is used to match entities through an attribute-level self-attention mechanism, and leverages the massive unlabeled data from new data sources through domain adaptation to make it generic and data-source agnostic.

Domain Adaptation Transfer Learning

Interrogating Paradigms in Self-supervised Graph Representation Learning

no code implementations29 Sep 2021 Puja Trivedi, Mark Heimann, Danai Koutra, Jayaraman J. Thiagarajan

Using the recent population augmentation graph-based analysis of self-supervised learning, we show theoretically that the success of GCL with popular augmentations is bounded by the graph edit distance between different classes.

Contrastive Learning Graph Representation Learning +2

How does Heterophily Impact the Robustness of Graph Neural Networks? Theoretical Connections and Practical Implications

1 code implementation14 Jun 2021 Jiong Zhu, Junchen Jin, Donald Loveland, Michael T. Schaub, Danai Koutra

We bridge two research directions on graph neural networks (GNNs), by formalizing the relation between heterophily of node labels (i. e., connected nodes tend to have dissimilar labels) and the robustness of GNNs to adversarial attacks.

Relational World Knowledge Representation in Contextual Language Models: A Review

no code implementations EMNLP 2021 Tara Safavi, Danai Koutra

Relational knowledge bases (KBs) are commonly used to represent world knowledge in machines.

Node Proximity Is All You Need: Unified Structural and Positional Node and Graph Embedding

1 code implementation26 Feb 2021 Jing Zhu, Xingyu Lu, Mark Heimann, Danai Koutra

While most network embedding techniques model the relative positions of nodes in a network, recently there has been significant interest in structural embeddings that model node role equivalences, irrespective of their distances to any specific nodes.

Graph Embedding Network Embedding

A Hidden Challenge of Link Prediction: Which Pairs to Check?

1 code implementation15 Feb 2021 Caleb Belth, Alican Büyükçakır, Danai Koutra

Thus, link prediction methods, which often rely on proximity-preserving embeddings or heuristic notions of node similarity, face a vast search space, with many pairs that are in close proximity, but that should not be linked.

Link Prediction Representation Learning

How do Quadratic Regularizers Prevent Catastrophic Forgetting: The Role of Interpolation

2 code implementations4 Feb 2021 Ekdeep Singh Lubana, Puja Trivedi, Danai Koutra, Robert P. Dick

Catastrophic forgetting undermines the effectiveness of deep neural networks (DNNs) in scenarios such as continual learning and lifelong learning.

Continual Learning

Towards Understanding and Evaluating Structural Node Embeddings

1 code implementation14 Jan 2021 Junchen Jin, Mark Heimann, Di Jin, Danai Koutra

While most network embedding techniques model the proximity between nodes in a network, recently there has been significant interest in structural embeddings that are based on node equivalences, a notion rooted in sociology: equivalences or positions are collections of nodes that have similar roles--i. e., similar functions, ties or interactions with nodes in other positions--irrespective of their distance or reachability in the network.

Network Embedding Social and Information Networks

Graph Neural Networks with Heterophily

1 code implementation28 Sep 2020 Jiong Zhu, Ryan A. Rossi, Anup Rao, Tung Mai, Nedim Lipka, Nesreen K. Ahmed, Danai Koutra

Graph Neural Networks (GNNs) have proven to be useful for many different practical applications.

From Static to Dynamic Node Embeddings

no code implementations21 Sep 2020 Di Jin, Sungchul Kim, Ryan A. Rossi, Danai Koutra

While previous work on dynamic modeling and embedding has focused on representing a stream of timestamped edges using a time-series of graphs based on a specific time-scale (e. g., 1 month), we propose the notion of an $\epsilon$-graph time-series that uses a fixed number of edges for each graph, and show its superiority over the time-scale representation used in previous work.

Time Series Time Series Analysis

CoDEx: A Comprehensive Knowledge Graph Completion Benchmark

2 code implementations EMNLP 2020 Tara Safavi, Danai Koutra

We present CoDEx, a set of knowledge graph completion datasets extracted from Wikidata and Wikipedia that improve upon existing knowledge graph completion benchmarks in scope and level of difficulty.

Benchmarking Link Prediction +1

G-CREWE: Graph CompREssion With Embedding for Network Alignment

1 code implementation30 Jul 2020 Kyle K. Qin, Flora D. Salim, Yongli Ren, Wei Shao, Mark Heimann, Danai Koutra

In this paper, we propose a framework, called G-CREWE (Graph CompREssion With Embedding) to solve the network alignment problem.

Mining Persistent Activity in Continually Evolving Networks

1 code implementation27 Jun 2020 Caleb Belth, Xinyi Zheng, Danai Koutra

Frequent pattern mining is a key area of study that gives insights into the structure and dynamics of evolving networks, such as social or road networks.

Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs

4 code implementations NeurIPS 2020 Jiong Zhu, Yujun Yan, Lingxiao Zhao, Mark Heimann, Leman Akoglu, Danai Koutra

We investigate the representation power of graph neural networks in the semi-supervised node classification task under heterophily or low homophily, i. e., in networks where connected nodes may have different class labels and dissimilar features.

Node Classification on Non-Homophilic (Heterophilic) Graphs

Neural Execution Engines: Learning to Execute Subroutines

1 code implementation NeurIPS 2020 Yujun Yan, Kevin Swersky, Danai Koutra, Parthasarathy Ranganathan, Milad Hashemi

A significant effort has been made to train neural networks that replicate algorithmic reasoning, but they often fail to learn the abstract concepts underlying these algorithms.

Learning to Execute

CONE-Align: Consistent Network Alignment with Proximity-Preserving Node Embedding

no code implementations10 May 2020 Xiyuan Chen, Mark Heimann, Fatemeh Vahedian, Danai Koutra

Network alignment, the process of finding correspondences between nodes in different graphs, has many scientific and industrial applications.

Evaluating the Calibration of Knowledge Graph Embeddings for Trustworthy Link Prediction

no code implementations EMNLP 2020 Tara Safavi, Danai Koutra, Edgar Meij

We first conduct an evaluation under the standard closed-world assumption (CWA), in which predicted triples not already in the knowledge graph are considered false, and show that existing calibration techniques are effective for KGE under this common but narrow assumption.

Knowledge Graph Embedding Knowledge Graph Embeddings +1

What is Normal, What is Strange, and What is Missing in a Knowledge Graph: Unified Characterization via Inductive Summarization

1 code implementation23 Mar 2020 Caleb Belth, Xinyi Zheng, Jilles Vreeken, Danai Koutra

We apply our rules to three large KGs (NELL, DBpedia, and Yago), and tasks such as compression, various types of error detection, and identification of incomplete information.

Knowledge Graphs Question Answering


no code implementations ICLR 2020 Yujun Yan, Kevin Swersky, Danai Koutra, Parthasarathy Ranganathan, Milad Hashemi

Turing complete computation and reasoning are often regarded as necessary pre- cursors to general intelligence.

On Proximity and Structural Role-based Embeddings in Networks: Misconceptions, Techniques, and Applications

no code implementations22 Aug 2019 Ryan A. Rossi, Di Jin, Sungchul Kim, Nesreen K. Ahmed, Danai Koutra, John Boaz Lee

Unfortunately, recent work has sometimes confused the notion of structural roles and communities (based on proximity) leading to misleading or incorrect claims about the capabilities of network embedding methods.

Misconceptions Network Embedding

node2bits: Compact Time- and Attribute-aware Node Representations for User Stitching

1 code implementation18 Apr 2019 Di Jin, Mark Heimann, Ryan Rossi, Danai Koutra

Identity stitching, the task of identifying and matching various online references (e. g., sessions over different devices and timespans) to the same user in real-world web services, is crucial for personalization and recommendations.


Latent Network Summarization: Bridging Network Embedding and Summarization

1 code implementation11 Nov 2018 Di Jin, Ryan Rossi, Danai Koutra, Eunyee Koh, Sungchul Kim, Anup Rao

Motivated by the computational and storage challenges that dense embeddings pose, we introduce the problem of latent network summarization that aims to learn a compact, latent representation of the graph structure with dimensionality that is independent of the input graph size (i. e., #nodes and #edges), while retaining the ability to derive node representations on the fly.

Social and Information Networks

t-PINE: Tensor-based Predictable and Interpretable Node Embeddings

no code implementations3 May 2018 Saba A. Al-Sayouri, Ekta Gujral, Danai Koutra, Evangelos E. Papalexakis, Sarah S. Lam

Contrary to baseline methods, which generally learn explicit graph representations by solely using an adjacency matrix, t-PINE avails a multi-view information graph, the adjacency matrix represents the first view, and a nearest neighbor adjacency, computed over the node features, is the second view, in order to learn explicit and implicit node representations, using the Canonical Polyadic (a. k. a.

General Classification Link Prediction +3

RECS: Robust Graph Embedding Using Connection Subgraphs

no code implementations3 May 2018 Saba A. Al-Sayouri, Danai Koutra, Evangelos E. Papalexakis, Sarah S. Lam

Representation learning algorithms aim to preserve local and global network structure by identifying node neighborhood notions.

General Classification Graph Embedding +5

REGAL: Representation Learning-based Graph Alignment

1 code implementation17 Feb 2018 Mark Heimann, Haoming Shen, Tara Safavi, Danai Koutra

Problems involving multiple networks are prevalent in many scientific and other domains.

Social and Information Networks

Driving with Data: Modeling and Forecasting Vehicle Fleet Maintenance in Detroit

1 code implementation18 Oct 2017 Josh Gardner, Danai Koutra, Jawad Mroueh, Victor Pang, Arya Farahi, Sam Krassenstein, Jared Webb

Understanding the existence of patterns and trends in this data could be useful to a variety of stakeholders, particularly as Detroit emerges from Chapter 9 bankruptcy, but the patterns in such data are often complex and multivariate and the city lacks dedicated resources for detailed analysis of this data.

Computers and Society

Graph Summarization Methods and Applications: A Survey

no code implementations14 Dec 2016 Yike Liu, Tara Safavi, Abhilash Dighe, Danai Koutra

While advances in computing resources have made processing enormous amounts of data possible, human ability to identify patterns in such data has not scaled accordingly.

Data Summarization

Linearized and Single-Pass Belief Propagation

1 code implementation27 Jun 2014 Wolfgang Gatterbauer, Stephan Günnemann, Danai Koutra, Christos Faloutsos

Often, we can answer such questions and label nodes in a network based on the labels of their neighbors and appropriate assumptions of homophily ("birds of a feather flock together") or heterophily ("opposites attract").

Graph-based Anomaly Detection and Description: A Survey

1 code implementation18 Apr 2014 Leman Akoglu, Hanghang Tong, Danai Koutra

This survey aims to provide a general, comprehensive, and structured overview of the state-of-the-art methods for anomaly detection in data represented as graphs.

Social and Information Networks Cryptography and Security

NetSimile: A Scalable Approach to Size-Independent Network Similarity

no code implementations12 Sep 2012 Michele Berlingerio, Danai Koutra, Tina Eliassi-Rad, Christos Faloutsos

Having such features will enable a wealth of graph mining tasks, including clustering, outlier detection, visualization, etc.

Social and Information Networks Physics and Society Applications

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