publications

2021

1. NeurIPS

   Adversarial Graph Augmentation to Improve Graph Contrastive Learning

   Suresh, Susheel, Li, Pan, Hao, Cong, and Neville, Jennifer

   In Thirty-fourth Conference on Neural Information Processing Systems 2021

   Abstract HTML PDF Code Talk Reviews

   Self-supervised learning of graph neural networks (GNN) is in great need because of the widespread label scarcity issue in real-world graph/network data. Graph contrastive learning (GCL), by training GNNs to maximize the correspondence between the representations of the same graph in its different augmented forms, may yield robust and transferable GNNs even without using labels. However, GNNs trained by traditional GCL often risk capturing redundant graph features and thus may be brittle and provide sub-par performance in downstream tasks. Here, we propose a novel principle, termed adversarial-GCL (AD-GCL), which enables GNNs to avoid capturing redundant information during the training by optimizing adversarial graph augmentation strategies used in GCL. We pair AD-GCL with theoretical explanations and design a practical instantiation based on trainable edge-dropping graph augmentation. We experimentally validate AD-GCL2 by comparing with the state-of-the-art GCL methods and achieve performance gains of up-to 14% in unsupervised, 6% in transfer, and 3% in semi-supervised learning settings overall with 18 different benchmark datasets for the tasks of molecule property regression and classification, and social network classification

2. KDD

   Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns

   Suresh, Susheel, Budde, Vinith, Neville, Jennifer, Li, Pan, and Ma, Jianzhu

   In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining 2021

   Abstract HTML PDF Code Talk Reviews

   Graph neural networks (GNNs) have achieved tremendous success on multiple graph-based learning tasks by fusing network structure and node features. Modern GNN models are built upon iterative aggregation of neighbor’s/proximity features by message passing. Its prediction performance has been shown to be strongly bounded by assortative mixing in the graph, a key property wherein nodes with similar attributes mix/connect with each other. We observe that real world networks exhibit heterogeneous or diverse mixing patterns and the conventional global measurement of assortativity, such as global assortativity coefficient, may not be a representative statistic in quantifying this mixing. We adopt a generalized concept, node-level assortativity, one that is based at the node level to better represent the diverse patterns and accurately quantify the learnability of GNNs. We find that the prediction performance of a wide range of GNN models is highly correlated with the node level assortativity. To break this limit, in this work, we focus on transforming the input graph into a computation graph which contains both proximity and structural information as distinct type of edges. The resulted multi-relational graph has an enhanced level of assortativity and, more importantly, preserves rich information from the original graph. We then propose to run GNNs on this computation graph and show that adaptively choosing between structure and proximity leads to improved performance under diverse mixing. Empirically, we show the benefits of adopting our transformation framework for semi-supervised node classification task on a variety of real world graph learning benchmarks.

2020

1. ICDM

   A Hybrid Model for Learning Embeddings and Logical Rules Simultaneously from Knowledge Graphs

   Suresh, Susheel, and Neville, Jennifer

   In 2020 IEEE International Conference on Data Mining (ICDM) 2020

   Abstract HTML PDF Talk Reviews

   The problem of knowledge graph (KG) reasoning has been widely explored by traditional rule-based systems and more recently by knowledge graph embedding methods. While logical rules can capture deterministic behavior in a KG, they are brittle and mining ones that infer facts beyond the known KG is challenging. Probabilistic embedding methods are effective in capturing global soft statistical tendencies and reasoning with them is computationally efficient. While embedding representations learned from rich training data are expressive, incompleteness and sparsity in real-world KGs can impact their effectiveness. We aim to leverage the complementary properties of both methods to develop a hybrid model that learns both high-quality rules and embeddings simultaneously. Our method uses a cross feedback paradigm wherein an embedding model is used to guide the search of a rule mining system to mine rules and infer new facts. These new facts are sampled and further used to refine the embedding model. Experiments on multiple benchmark datasets show the effectiveness of our method over other competitive standalone and hybrid baselines. We also show its efficacy in a sparse KG setting.

2018

1. AAAI

   VoC-DL: Revisiting Voice Of Customer Using Deep Learning

   Suresh, Susheel, S, Guru Rajan T, and Gopinath, Vipin

   In AAAI Conference on Artificial Intelligence 2018

   Abstract HTML PDF Reviews

   In the field of digital marketing, understanding the voice of the customer is paramount. Mining textual content written by visitors on websites or social media can offer new dimensions to marketers and CX executives. Traditional tasks in NLP like sentiment analysis, topic modeling etc. can solve only certain specific problems but don’t provide a generic solution to identifying/understanding the intention behind a text. In this paper we consider higher dimensional extensions to the sentiment concept by incorporating labels like product enquiry, buying intent, seeking help, feedback and pricing query which give us a deeper understanding of the text. We show how our model performs in a real-world enterprise use case. Word2Vec embeddings are used for word representations and later we compare three algorithms for classification. SVM’s provide us with a strong baseline. Two deep learning models viz. vanilla CNN and RNN’s with LSTM are compared. With no use of hard-coded or hand engineered features, our generic model can be used in a variety of use cases where text mining is involved with ease.

2016

1. ICVGIP

   Towards Quantifying the Amount of Uncollected Garbage through Image Analysis

   Suresh, Susheel, Sharma, Tarun, K., Prashanth T., V, Subramaniam, Sitaram, Dinkar, and M, Nirupama

   In Proceedings of the Tenth Indian Conference on Computer Vision, Graphics and Image Processing 2016

   Abstract HTML PDF Code Reviews

   Civic authorities in many Indian cities have a tough time in garbage collection and as a result there is a pile up of garbage in the cities. In order to manage the situation, it is first required to be able to quantify the issue. In this paper, we address the problem of quantification of garbage in a dump using a two step approach. In the first step, we build a mobile application that allows citizens to capture images of garbage and upload them to a server. In the second step, back-end performs analysis on these images to estimate the amount of garbage using computer vision techniques. Our approach to volume estimation uses multiple images of the same dump (provided by the mobile application) from different perspectives, segments the dump from the background, reconstructs a three dimensional view of the dump and then estimates its volume. Using our novel pipeline, our experiments indicate that with 8 different perspectives, we are able to achieve an accuracy of about 85 % for estimating the volume.

2015

1. ICISS

   Enforcing Secure Data Sharing in Web Application Development Frameworks Like Django Through Information Flow Control

   Suresh, Susheel, Narendra Kumar, N. V., and Shyamasundar, R. K.

   In Information Systems Security 2015

   Abstract HTML PDF

   The primary aim of web application development frameworks like Django is to provide a platform for developers to realize applications from concepts to launch as quickly as possible. While Django framework provides hooks that enable the developer to avoid the common security mistakes, there is no systematic way to assure compliance of a security policy while developing an application from various components. In this paper, we show the security flaws that arise by considering different versions of an application package and then show how, these mistakes that arise due to incorrect flow of information can be overcome using the Readers-Writers Flow Model that has the ability to manage the release and subsequent propagation of information.