Amit Dhurandhar  Amit Dhurandhar photo       

contact information

Research Scientist - machine learning, data mining
Thomas J. Watson Research Center, Yorktown Heights, NY USA
  +1dash914dash945dash1325

links


more information

More information:  Resume  |  Research Statement



Patents

 

    • Amit Dhurandhar, Sechan Oh and Marek Petrik. Interpretable Rule Generation using loss-preserving transformation. U.S. Provisional Patent Application Serial No. 15/489,418
    • Guillermo Cecchi,Amit Dhurandhar, Stacey M Gifford, Raquel Norel, Pablo Meyer rojas, Kahn Rhrissorrakrai and Bo Zhang. Predicting User Preferences based on Olfactory Characteristics. YOR8020161333
    • Ioana Baldini, Amit Dhurandhar, Abhishek Kumar, Aleksandra Mojsilovic, Kein T Pham, Kush R Varshney andMaja Vukovic. Humanitarian Crisis analysis using secondary information gathered by focused web crawler. YOR920161631
    • Guillermo Cecchi, Amit Dhurandhar and Pablo Meyer. Correlating Olfactory Perception with Molecular Structure. YOR920161332
    • Amit Dhurandhar, Bruce Graves, Rajesh Ravi and Markus Ettl. A System and Method for Identifying Procurement Fraud/Risk. U.S. Provisional Patent Application Serial No. 14/186,071
    • Amit Dhurandhar, Stuart Seigal, Yada Zhu and Jayant Kalagnanam. A System and Method for Detecting Electricity Theft via Meter Tampering Using Statistical Methods of Anomaly Detection. U.S. Provisional Patent Application Serial No. 13/909,239
    • Amit Dhurandhar and Jun Wang. A System and Method for Relational Transductive Learning. U.S. Provisional Patent Application Serial No. 13/787,807 (granted)
    • Pawan Chowdhary, Amit Dhurandhar, Markus Ettl, Soumyadip Ghosh, Bruce Graves, Bill Schaefer and Yu Tang. Method and system for optimizing procurement spend compliance. U.S. Provisional Patent Application Serial No. 13/339,626 (granted)
    • Robert Baseman, Amit Dhurandhar, Sholom Weiss and Brian White. A System and Method for Continuous Prediction of Expected Chip Performance Throughout the Production Lifecycle. U.S. Provisional Patent Application Serial No. 13/242,692 (granted)
    • Amit Dhurandhar. Improving Predictions using Aggregate Information. U.S. Provisional Patent Application Serial No. 13/184,000 (granted)
    • Amit Dhurandhar and Jayant Kalagnanam. Multistep Time Series Prediction in Complex Instrumented Domains. U.S. Provisional Patent Application Serial No. 12/966,465 (granted)
    • Amit Dhurandhar, Robert Baseman and Fateh Tipu. A System and Method for Identifying Significant Consumable Insensitive Trace Features. YOR8020140228
    • Amit Dhurandhar, Bruce Graves, Rajesh Ravi, Gopikrishnan Maniachari, Markus Ettl, Anthony Mazzatti. A Robust System for Ranking and Tracking Suspicious Procurement Entities. YOR8020140315

 


Publications

 

Journals

        • Andreas Keller, Richard C. Gerkin, Yuanfang Guan, Amit Dhurandhar, Gabor Turu, Bence Szalai, Joel D.Mainland, Yusuke Ihara,Chung Wen Yu, Russ Wolfinger, Celine Vens, Leander Schietgat, Kurt De Grave, Raquel Norel, DREAM Olfaction Consortium, Gustavo Stolovitzky, Guillermo Cecchi, Leslie B. Vosshall, and Pablo Meyer. Predicting Human Olfactory Perception from Chemical Features of Odor Molecules. Science, 2017. article (Highlighted at AAAS meeting as a breakthrough in the last 3 decades in olfactory research)
        • Tsuyoshi Ide and Amit Dhurandhar. Supervised Item Response Models for Informative Prediction. Knowledge and Information Systems (KAIS), 2016. The final publication is available at official link and the personal version is here PDF (invited)
        • Amit Dhurandhar and Karthik Sankarnarayanan. Improving Classification Performance through Selective Instance Completion. Machine Learning Journal (MLJ), 2015. The final publication is available at official link and the personal version is here PDF (with presentation slot at ECML 2015)
        • Amit Dhurandhar and Marek Petrik. Efficient and Accurate Methods for Updating Generalized Linear Models with Multiple Feature Additions. Journal of Machine Learning Research (JMLR), 2014. PDF
        • Amit Dhurandhar. Bounds on the Moments for an Ensemble of Random Decision Trees. Knowledge and Information Systems (KAIS), 2014.The final publication is available at official link and the personal version is here PDF
        • Sholom Weiss, Amit Dhurandhar, Robert Baseman, Brian White, Ronald Logan, Jonathan Winslow and Daniel Poindexter. Continuous Prediction of Manufacturing Outcomes Throughout the Production Lifecycle. Journal of Intelligent Manufacturing (JIMS), 2014. The final publication is available at official link and the personal version is here PDF
        • Amit Dhurandhar and Jun Wang. Single Network Relational Transductive Learning. Journal of Artificial Intelligence Research (JAIR), 2013. PDF
        • Amit Dhurandhar. Using Coarse Information for Real Valued Prediction. Data Mining and Knowledge Discovery (DMKD), 2013. (nominated for IBM Pat Goldberg Award) The final publication is available at official link and the personal version is here PDF
        • Amit Dhurandhar and Alin Dobra. Probabilistic Characterization of Nearest Neighbor Classifiers. Intl. Journal of Machine Learning and Cybernetics (IJMLC), 2012. (invited) The final publication is available at official link and the personal version is here PDF
        • Amit Dhurandhar and Alin Dobra. Distribution free bounds for Relational Classification. Knowledge and Information Systems (KAIS), 2012. The final publication is available at official link and the personal version is here PDF
        • Amit Dhurandhar and Alin Dobra. Semi-analytical Method for Analyzing Models and Model Selection Measures based on Moment Analysis. ACM Transactions on Knowledge Discovery from Data (TKDD), 2009. PDF
        • Amit Dhurandhar and Alin Dobra. Probabilistic Characterization of Random Decision Trees. Journal of Machine Learning Research (JMLR), 2008. PDF
        • Amit Dhurandhar and Alin Dobra. Test Set Bounds for Relational Data that vary with Strength of Dependence. submitted PDF

 

Conferences\Workshops

 

        • Amit Dhurandhar, Vijay Iyengar, Ronny Luss and Karthikeyan Shanmugam. A Formal Framework to Characterize Interpretability of Procedures. Human Interpretability in Machine Learning workshop in Intl. Conference on Machine Learning (ICML), 2017. PDF
        • Amit Dhurandhar, Margareta Ackerman and Xiang Wang. Uncovering Group Level Insights with Accordant Clustering. SIAM Intl. Conference on Data Mining (SDM), 2017 (oral). PDF, supplementary material
        • Amit Dhurandhar, Sechan Oh and Marek Petrik. Building Interpretable Recommender via Loss-Preserving Transformation. Human Interpretability in Machine Learning workshop in Intl. Conference on Machine Learning (ICML), 2016. PDF
        • Amit Dhurandhar, Bruce Graves, Rajesh Ravi, Gopikrishnan Maniachari and Markus Ettl. Big Data System for Analyzing Risky Entities. ACM SIGKDD conference on Knowledge Discovery and Data Mining (KDD), 2015 (oral). PDF
        • Tsuyoshi Ide and Amit Dhurandhar. Informative Prediction based on Ordinal Questionnaire Data. IEEE Intl. Conference on Data Mining (ICDM), 2015 (Best paper candidate) PDF
        • Amit Dhurandhar, Rajesh Ravi, Bruce Graves, Gopikrishnan Maniachari and Markus Ettl. Robust System for Identifying Procurement Fraud. Innovative Applications of Artificial Intelligence track in Assoc. for Adv. in Artificial Intelligence (AAAI), 2015. (Deployed Application Award)
        • Amit Dhurandhar and Karthik Gurumoorthy. Symmetric Submodular Clustering with Actionable Constraint. Discrete Optimization workshop in, (NIPS) 2014. PDF
        • Rajesh Ravi, Amit Dhurandhar, Markus Ettl, Bruce Graves. Procurement Fraud Risk Analytics Tool. Information on Demand Conference (IOD), 2014.
        • Sholom Weiss, Amit Dhurandhar and Robert Baseman. Improving Quality Control by Early Prediction of Manufacturing Outcomes. ACM SIGKDD conference on Knowledge Discovery and Data Mining (KDD), 2013 (oral). PDF
        • Karthik Sankarnarayanan and Amit Dhurandhar. Intelligently Querying Incomplete Instances for Improving Classification Performance. ACM International Conference on Information and Knowledge Management (CIKM), 2013 (full paper, oral). PDF
        • Amit Dhurandhar. Auto-correlation Dependent Bounds for Relational Data. Mining and Learning over Graphs workshop in ACM SIGKDD conference on Knowledge Discovery and Data Mining (KDD), 2013. PDF
        • Amit Dhurandhar. Improving Predictions using Aggregate Information. ACM SIGKDD conference on Knowledge Discovery and Data Mining (KDD), 2011. PDF
        • Pawan Chowdhary, Markus Ettl, Amit Dhurandhar, Soumyadip Ghosh, Gopikrishna Maniachari, Bruce Graves, Bill Schaefer and Yu Tang. Identify and Manage Procurement Savings using Advanced Compliance Analytics. IEEE International Conference on e-Business Engineering (ICEBE), 2011. PDF
        • Amit Dhurandhar. Multistep Time Series Prediction in Complex Instrumented Domains. Large scale analytics in complex instrumented domains workshop in IEEE International Conference on Data Mining (ICDM), 2010. PDF     This paper was also invited to Chance Discovery workshop in (IJCAI), 2011.
        • Amit Dhurandhar. Learning Maximum Lag for Grouped Graphical Granger Models. Knowledge Discovery from Climate Data Prediction, Extremes, and Impacts workshop in IEEE International Conference on Data Mining (ICDM), 2010. PDF
        • Dan Connors, Amit Dhurandhar, Markus Ettl, Mary Helander, Jayant Kalagnanam, Shubir Kapoor, Ramesh Natarajan, Stuart Seigal, Zhackary Xue. Demand forecasting and supply chain optimization using freshness. Information on Demand Conference (IOD), 2010.
        • Robert Baseman, Amit Dhurandhar, Michal Ozery and Naama Perush. Statistical Assessment of dissimilarities in trace data of unusual and nominal wafers. ISMI manufacturing week, 2010.
        • Robert Baseman, Frances Clougherty, Amit Dhurandhar, Lyndon Logan, Daniel Poindexter, Brian White, Sholom Weiss, Jonathan Winslow, Denis Zhereschin. Early Predictions of Device Performance for Enhanced Process Control and Operations Optimization. ISMI Symposium on manufacturing excellence, 2010.
        • John Andrews, Robert Baseman, Michael Biagetti, Amit Dhurandhar, Hong Lin, Michal Ozery-Flato, Stuart A Siegel, Naama Parush-Shear-Yashuv, Adam Ticknor. Utilization of Equipment Trace Data in a 300mm Semiconductor Fab. ISMI Symposium on manufacturing excellence, 2010.
        • Amit Dhurandhar and Alin Dobra. Evaluating Evaluation Measures. Evaluation Methods in Machine Learning workshop in International Conference on Machine Learning (ICML), 2009. PDF
        • Amit Dhurandhar and Alin Dobra. Study of Classification Algorithms using Moment Analysis. One of 2 regular papers accepted to New Challenges in Theoretical Machine Learning workshop in Neural Information Processing Systems (NIPS), 2008. PDF Talk link
        • Amit Dhurandhar, Kartik Shankar and Rakesh Jawale. Robust Pattern Recognition Scheme for Devanagari Script. IEEE International Conference on Computational Intelligence and Security (CIS) 2005.

 

Technical Reports

 

        • Amit Dhurandhar and Paul Gader. Output Distribution of Choquet Integral. PDF
        • Amit Dhurandhar and Alin Dobra. Insights into Cross-validation. PDF
        • Amit Dhurandhar and Alin Dobra. Independent vs Collective Classification in Statistical Relational Learning. submitted PDF