Nathan Hodas
Nathan Hodas
Biography
Nathan Hodas arrived at Pacific Northwest National Laboratory (PNNL) in 2014, after a postdoc at the Information Sciences Institute at University of Southern California. His PhD focused on experimental and theoretic biophysics and nanoscale non-linear optics. At the Information Sciences Institute he studied social network dynamics and developed human response dynamics approach to characterizing information spread.
Hodas's research at PNNL revolves around the use of data analytics and machine learning to advance scientific research. His current research focuses on leveraging few-shot learning and human-computer interaction to support applications of artificial intelligence. His other projects include using deep learning to learn communication patterns, modeling intent and adversarial behavior, and applying machine learning to scientific problems.
Disciplines and Skills
- Analytics
- Artificial Intelligence
- Biophysics
- Data Science
- Deep Learning
- Human Computer Interaction
- Machine Learning
- Nonlinear Optics
Education
- PhD in physics, California Institute of Technology
- BA in physics, Williams College
Awards and Recognitions
- Ronald L. Brodzinski Early Career Exceptional Achievement Award, 2017
- Author of the Year, National Security Directorate, PNNL, 2017
- APS Leroy Apker Award, 2004
Publications
2021
- Kvinge H.J., Z.D. New, N.C. Courts, J.H. Lee, L.A. Phillips, C.D. Corley, and A.R. Tuor, et al. 2021. "Fuzzy Simplicial Networks: A Topology-Inspired Model to Improve Task Generalization in Few-shot Learning." In AAAI Workshop on Meta-Learning and MetaDL Challenge, February 9, 2021, Virtual, Online. Proceedings of Machine Learning Research, edited by I. Guyon, et al, 140, 77-89. Brookline, Massachusetts:ML Research Press. PNNL-SA-158201.
- Kvinge H.J., Z.D. New, N.C. Courts, J.H. Lee, L.A. Phillips, C.D. Corley, and A.R. Tuor, et al. 2021. "Fuzzy Simplicial Networks: A Topology-Inspired Model to Improve Task Generalization in Few-shot Learning." In Thirty-Fifth AAAI Conference on Artificial Intelligence. PNNL-SA-156013.
- Fujimoto T.C., T.J. Doster, A. Attarian, J.M. Brandenberger, and N.O. Hodas. 2021. "The Effect of Antagonistic Behavior in Reinforcement Learning." In AAAI-21 Workshop on Reinforcement Learning in Games, February 8, 2021, Virtual. Menlo Park, California:Association for the Advancement of Artificial Intelligence. PNNL-SA-157666.
- Fujimoto T.C., T.J. Doster, A. Attarian, J.M. Brandenberger, and N.O. Hodas. 2021. "Reward-Free Attacks in Multi-Agent Reinforcement Learning." In Reinforcement Learning for Real Life Workshop @ ICML 2021. PNNL-SA-162992.
- Xu W., K. Balaguru, A. August, N. Lalo, N.O. Hodas, M. DeMaria, and D.R. Judi. 2021. "Deep Learning Experiments for Tropical Cyclone Intensity Forecasts." Weather and Forecasting 36, no. 4:1453–1470. PNNL-SA-153486. doi:10.1175/WAF-D-20-0104.1
- Girard M.K., A.R. Hagen, I.J. Schwerdt, M.E. Gaumer, L.W. McDonald, N.O. Hodas, and E.R. Jurrus. 2021. "Uranium Oxide Synthetic Pathway Discernment through Unsupervised Morphological Analysis." Journal of Nuclear Materials 552. PNNL-SA-156254. doi:10.1016/j.jnucmat.2021.152983
- Fujimoto T.C., T.J. Doster, A. Attarian, J.M. Brandenberger, and N.O. Hodas. 2021. "Reward-Free Attacks in Multi-Agent Reinforcement Learning." In Learning in Presence of Strategic Behavior (NeurIPS 2021 Workshop). PNNL-SA-168665.
2020
- Colby S.M., J. Nunez, N.O. Hodas, C.D. Corley, and R.S. Renslow. 2020. "Deep learning to generate in silico chemical property libraries and candidate molecules for small molecule identification in complex samples." Analytical Chemistry 92, no. 2:1720-1729. PNNL-SA-144150. doi:10.1021/acs.analchem.9b02348
- Phillips L.A., Z.D. New, A. Avila, E.R. Burtner, A.J. Kingsland, H.J. Kvinge, and N.D. O'Brien, et al. 2020. "Sharkzor: Human-in-the-loop Deep Learning for Image Classification from Few Examples." Journal of Intelligence Community Research and Development. PNNL-SA-151403.
- Skomski S.E., A.R. Tuor, A. Avila, L.A. Phillips, Z.D. New, H.J. Kvinge, and C.D. Corley, et al. 12/11/2020. "Prototypical Region Proposal Networks for Few-Shot Localization and Classification." Presented by S.E. Skomski at NeurIPS 2020 MetaLearn Workshop, Online, Canada. PNNL-SA-158132.
- Skomski S.E., A.R. Tuor, A. Avila, L.A. Phillips, Z.D. New, H.J. Kvinge, and C.D. Corley, et al. 2020. "Prototypical Region Proposal Networks for Few-Shot Localization and Classification." In 4th Workshop on Meta-Learning at NeurIPS 2020, December 11, 2020, (Virtual Only). San Diego, California:Neural Information Processing Systems. PNNL-SA-158131
2019
- Saldanha E.G., L.M. Blaha, A. Visweswara Sathanur, N.O. Hodas, S. Volkova, and M.T. Greaves. 2019. "Evaluation and Validation Approaches for Simulation of Social Behavior: Challenges and Opportunities." In Social-Behavioral Modeling for Complex Systems, edited by Paul K. Davis, Angela O'Mahony, Jonathan Pfautz. 495-515. Hoboken, New Jersey:John Wiley & Sons, Inc. PNNL-SA-136082. doi:10.1002/9781119485001.ch21
- Yeung E.H., S. Kundu, and N.O. Hodas. 2019. "Learning Deep Neural Network Representations for Koopman Operators of Nonlinear Dynamical Systems." In Proceedings of the American Control Conference, July 10-12, 2019, Philadelphia, PA, 4832-4839; Article number 8815339. Piscataway, New Jersey:IEEE. PNNL-SA-128573.
2018
- Hilliard N.C., N.C. Hilliard, L.A. Phillips, S.A. Howland, A. Yankov, C.D. Corley, and N.O. Hodas. 2018. "Few-Shot Learning with Metric-Agnostic Conditional Embeddings." In arXiv. PNNL-SA-140755.
- Coleman A., J.M. Brandenberger, J.D. Tagestad, D.R. Judi, N.O. Hodas, K.B. Larson, and B.S. Kravitz, et al. 2018. Next Generation Geospatial Analytics Summit Report: September 2017. PNNL-27358. Richland, WA: Pacific Northwest National Laboratory.
- Goh G.B., C.M. Siegel, A. Vishnu, N.O. Hodas, and N.A. Baker. 2018. "How Much Chemistry Does a Deep Neural Network Need to Know to Make Accurate Predictions?." In IEEE Winter Conference on Applications of Computer Vision (WACV 2018), March 12-15, 2018, Lake Tahoe, NV, 1340-1349. Piscataway, New Jersey:IEEE. PNNL-SA-127201. doi:10.1109/WACV.2018.00151
- Lee J., L.M. Bramer, N.O. Hodas, C.D. Corley, and S.H. Payne. 06/06/2018. "Deep learning benchmark data for de novo peptide sequencing." Abstract submitted to Machine Learning in Science and Engineering, Pittsburgh, Pennsylvania. PNNL-SA-134053.
- Hodas N.O., and P. Stinis. 2018. "Doing the impossible: Why neural networks can be trained at all." Frontiers in Psychology 9, no. JUN:Article No. 1185. PNNL-SA-127608. doi:10.3389/fpsyg.2018.01185
- Phillips L.A., G.B. Goh, and N.O. Hodas. 2018. "Explanatory Masks for Neural Network Interpretability." In ICML/IJCAI Workshop on Explainable Artificial Intelligence (XAI-18). PNNL-SA-132289.
- Goh G.B., C.M. Siegel, A. Vishnu, and N.O. Hodas. 2018. "Using Rule-Based Models for Weak Supervised Learning: A ChemNet for Transferable Chemical Property Prediction." In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD 2018), August, 2018, London, UK, 302-306. New York, New York:Association for Computing Machinery. PNNL-SA-132274. doi:10.1145/3219819.3219838
- Corley C.D., N.O. Hodas, E.H. Yeung, A.M. Tartakovsky, T.J. Hagge, S. Choudhury, and K. Agarwal, et al. 2018. "Deep Learning for Scientific Discovery." The Next Wave 22, no. 1:27-31. PNNL-SA-129480.
- Bakker C., M.J. Henry, and N.O. Hodas. 2018. "The Outer Product Structure of Neural Network Derivatives." arXiv. [Preprint]. Submitted October 09, 2018. PNNL-SA-138611. doi:10.48550/arXiv.1810.03798
2017
- Goh G.B., N.O. Hodas, and A. Vishnu. 2017. "Deep Learning for Computational Chemistry." Journal of Computational Chemistry 38, no. 16:1291-1307. PNNL-SA-121040. doi:10.1002/jcc.24764
- Coleman A., J.M. Brandenberger, J.D. Tagestad, D.R. Judi, N.O. Hodas, K.B. Larson, and B.S. Kravitz, et al. 2018. Next Generation Geospatial Analytics Summit Report: September 2017. PNNL-27358. Richland, WA: Pacific Northwest National Laboratory.
- Hohman F.M., N.O. Hodas, and D. Chau. 2017. "ShapeShop: Towards Understanding Deep Learning Representations via Interactive Experimentation." In Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems (CHI EA 2017), May 6-11, 2017, Denver, Colorado, 1694-1699. New York, New York:ACM. PNNL-SA-126087. doi:10.1145/3027063.3053103
- Phillips L.A., and N.O. Hodas. 2017. "Assessing the Linguistic Productivity of Unsupervised Deep Neural Networks." In Proceedings of the 39th Annual Meeting of the Cognitive Sciencey Society (CogSci 2017), July 26-29, 2017, London, United Kingdom, 937-942. Austin, Texas:Cognitive Science Society. PNNL-SA-126047.
- Volkova S., K.J. Shaffer, J. Jang, and N.O. Hodas. 2017. "Separating Facts from Fiction: Linguistic Models to Classify Suspicious and Trusted News Posts on Twitter." In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, July 30-August 4, 2017, Vancouver, BC, Canada, 2, 647-653; Paper No. 10.18653/v1/P17-2102. PNNL-SA-123856. doi:10.18653/v1/P17-2102
- Phillips L.A., K.J. Shaffer, D.L. Arendt, N.O. Hodas, and S. Volkova. 2017. "Intrinsic and Extrinsic Evaluation of Spatiotemporal Text Representations in Twitter Streams." In Proceedings of the 2nd Workshop on Representation Learning for NL (Rep4NLP@ACL 2017), August 3, 2017, Vancouver Canada, 201-210. Stroudsburg, Pennsylvania:Association for Computational Linguistic. PNNL-SA-122225.
- Pirrung M.A., N.C. Hilliard, A. Yankov, N.D. O'Brien, P.J. Weidert, C.D. Corley, and N.O. Hodas. 2017. "Sharkzor: Interactive Deep Learning for Image Triage, Sort and Summary." In Human in the Loop Machine Learning Workshop at the International Conference on Machine Learning (ICML). PNNL-SA-126641.
- Hilliard N.C., N.O. Hodas, and C.D. Corley. 2017. "Dynamic Input Structure and Network Assembly for Few-Shot Learning." In ICML 2017 AutoML Workshop, JMLR: Workshop and Conference. PNNL-SA-127672.
- Hilliard N.C., N.O. Hodas, and C.D. Corley. 08/06/2017. "Dynamic Input Structure and Network Assembly for Few-Shot Learning." Presented by Nathan O Hodas at ICML AutoML Workshop, Sydney, Australia. PNNL-SA-128053.
- Goh G.B., N.O. Hodas, C.M. Siegel, and A. Vishnu. 2017. "SMILES2Vec: An Interpretable General-Purpose Deep Neural Network for Predicting Properties from Chemical Text Representations." In ICMLA. PNNL-SA-129036.
- Goh G.B., C.M. Siegel, A. Vishnu, and N.O. Hodas. 2017. "ChemNet: A Transferable and Generalizable Deep Neural Network for Small-Molecule Property Prediction." In Machine Learning for Molecules and Materials (NIPS 2017 Workshop), December 8, 2017, Long Beach, California. La Jolla, California:Neural Information Processing Systems Foundation, Inc. PNNL-SA-129942.
- Hodas N.O., K.J. Shaffer, A. Yankov, C.D. Corley, and A.L. Anderson. 2017. "Beyond Fine Tuning: Adding capacity to leverage few labels." In Learning with Limited Labeled Data (LLD Workshop 2017), December 9, 2017, Long Beach, California. La Jolla, California:Neural Information Processing Systems Foundation, Inc. PNNL-SA-122155.
- Corley C.D., and N.O. Hodas. 2017. "Cognitive Modeling of the Impact of Wireless Emergency Alerts." In The Future of Emergency Alert and Warning Systems: Research Directions. PNNL-SA-126197.
2016
- Jurrus E.R., N.O. Hodas, N.A. Baker, T.P. Marrinan, and M.D. Hoover. 2016. "Adaptive Visual Sort and Summary of Micrographic Images of Nanoparticles for Forensic Analysis." In 2016 IEEE International Symposium on Technologies for Homeland Security, May 10-11, 2016, Waltham, MA. Piscataway, New Jersey:IEEE. PNNL-SA-114724. doi:10.1109/THS.2016.7568959
- Hunter J.S., and N.O. Hodas. 2016. "Mutual Information for fitting deep nonlinear models." arXiv. [Preprint]. Submitted December 16, 2016. PNNL-SA-121434. doi:10.48550/arXiv.1612.05708
2015
- Corley C.D., N.O. Hodas, and S. Lo. 2015. Training Datasets of Risk Communication and Perception: Task Final Report. PNNL-24113. Richland, WA: Pacific Northwest National Laboratory.
- Hodas N.O., G. Ver Steeg, J.J. Harrison, S. Chikkagoudar, E.B. Bell, and C.D. Corley. 2015. "Disentangling the Lexicons of Disaster Response in Twitter." In Proceedings of the 24th International Conference on World Wide Web (WWW 2015), May 18-22, 2015, Florence, Italy, 1201-1204. New York, New York:ACM. PNNL-SA-103652. doi:10.1145/2740908.2741728