Temporal Energy Transformer for Long Range Propagation in Continuous Time Dynamic Graphs
PSA Salam, A Abhishek, H Pandey
TMLR • 2026
energy transformercontinuous time dynamic graphstemporal graph
About
Hello! I am a machine learning researcher passionate about neural networks who enjoys working at the intersection of research and product.
I earned my M.S. (by Research) from the Indian Institute of Technology, Delhi. I work on projects that push boundaries and solve real-world problems. When I'm not working, you can find me running, swimming, or playing football or badminton.
Publications
Gnnxemplar: Exemplars to explanations-natural language rules for global GNN interpretability
B Armgaan, E Jain, H Pandey, M Chandran, S Ranu
NeurIPS (Oral) • 2025
GNN ExplainabilityNatural Language ExplanationExplainability
Data-Dependent Generalization Bounds for Neural Networks with ReLU
H Pandey, A Bagchi, SJ Bedathur, A Bhattacharya
TMLR • 2024
Generalization BoundStabilityDeep Learning Theory
Machine learning-based prediction of heavy metal leaching efficiency from zinc metallurgical waste
Partha Pratim Mondal, Harsh Panday, Bradley Martin Guy, Soniya Dhiman, Shaikh Ziauddin Ahammad, Rohan Jain
Chemical Engineering Journal: Green and Sustainable • 2026
Hydrometallurgical leachingMachine learningCritical metalsGallium and germanium recovery
Projects
Scalable Graph Explainability for Real-World Fraud Detection
FeaturedA scalable heterogeneous GNN explainability algorithm for real-world fraud detection on graphs with over 150 million nodes.
graph-neural-networksgraph-explainabilityfraud-detectionheterogeneous-graphsexplainable-ai
GNNXemplar: Exemplars to Explanations
FeaturedA global GNN explainer that uses LLMs to generate natural-language rules for interpreting Graph Neural Networks.
graph-neural-networksllmsgraph-explainabilityexplainable-ainatural-language
Temporal Energy Transformer
An energy-based graph attention model for long-range propagation in continuous-time dynamic graphs.
temporal-graphsgraph-neural-networksgraph-attentiontransformersdeep-learningenergy based model
Data-Dependent Generalization Bounds for Neural Networks
FeaturedData-dependent generalization bounds for ReLU neural networks establishing a connection between convergence and generalization.
generalizationdeep-learning-theorylearning-theoryneural-networksrelu
POND
A fast binary classifier combining gradient-based learning with tree-like efficiency for classification in scarce-data regimes.
machine-learningbinary-classificationtabular-datagradient-based-learningstability
Skills
Programming Languages
PythonC/C++JavaSQLJavaScriptHTML/CSS
Machine Learning & Deep Learning
PyTorchTensorFlowscikit-learn
Data & Visualization
PandasNumPyMatplotlib
Web & Application Frameworks
FlaskStreamlitReactHTML/CSSJavaScript
Developer Tools
GitDockerVimVS Code
Cloud & Platforms
Google Cloud PlatformLinux
Life
2024-03-15
Weekend Hike
A beautiful weekend hike in the mountains. The weather was perfect and the views were breathtaking.
## Experience
Spent the day exploring trails and enjoying nature. It was a great way to recharge a...
outdoorhikingnature