Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLP/CoNLL), 2007. Training Classifiers with Natural Language Explanations. Their, This "Cited by" count includes citations to the following articles in Scholar. Liu, E., Haghgoo, B., Chen, A. S., Raghunathan, A., Koh, P., Sagawa, S., Liang, P., Finn, C., Meila, M., Zhang, T. Catformer: Designing Stable Transformers via Sensitivity Analysis. His two research goals are (i) to make machine learning more robust, fair, and interpretable; and (ii) to make computers easier to communicate with through natural language. View details for DOI 10.1097/FJC.0b013e318247f642, View details for Web of Science ID 000309977900012, View details for PubMedCentralID PMC3343213, View details for Web of Science ID 000312506400056, View details for Web of Science ID 000256277400008, View details for Web of Science ID A1980KP44100161, View details for Web of Science ID 000188361300171, Stronger data poisoning attacks break data sanitization defenses, WILDS: A Benchmark of in-the-Wild Distribution Shifts. from MIT, 2004; Ph.D. from UC Berkeley, 2011). from MIT, 2004; Ph.D. from UC Berkeley, 2011). MI #~__ Q$.R$sg%f,a6GTLEQ!/B)EogEA?l kJ^- \?l{ P&d\EAt{6~/fJq2bFn6g0O"yD|TyED0Ok-\~[`|4P,w\A8vD$+)%@P4 0L ` ,\@2R 4f Associate Professor of Computer Science, Stanford University - Cited by 38,800 - machine learning - natural language processing . Liu, E., Raghunathan, A., Liang, P., Finn, C., Meila, M., Zhang, T. Just Train Twice: Improving Group Robustness without Training Group Information. Our model represents each individual's features over time as a nonlinear function of a low-dimensional, linearly-evolving latent state. Try again later. Percy Liang is an Associate Professor of Computer Science at Stanford University (B.S. He, H., Balakrishnan, A., Eric, M., Liang, P., Barzilay, R., Kan, M. Y. Naturalizing a Programming Language via Interactive Learning. Long, R., Pasupat, P., Liang, P., Erk, K., Smith, N. A. Pasupat, P., Liang, P., Erk, K., Smith, N. A. Textbook: Yes. His research spans many topics in machine learning and natural language processing, including robustness, interpretability, semantics, and reasoning. 390Jane Stanford Way Bouchard-Ct, A., Liang, P., Griffiths, T., Klein, D. Liang, P., Klein, D., Jordan, Michael, I. Professor Liang writes code faster than anyone I've ever seen. Feature Noise Induces Loss Discrepancy Across Groups. His awards include the Presidential Early Career Award for Scientists and Engineers (2019), IJCAI Computers and Thought Award (2016), an NSF CAREER Award (2016), a Sloan Research Fellowship (2015), a Microsoft Research Faculty Fellowship (2014), and multiple paper awards at ACL, EMNLP, ICML, and COLT. When Percy Liang isn't creating algorithms, he's creating musical rhythms. How much of a hypertree can be captured by windmills? A Tight Analysis of Greedy Yields Subexponential Time Approximation for Uniform Decision Tree, Enabling Language Models to Fill in the Blanks, Donahue, C., Lee, M., Liang, P., Assoc Computat Linguist, ExpBERT: Representation Engineering with Natural Language Explanations, Murty, S., Koh, P., Liang, P., Assoc Computat Linguist, Pretraining deep learning molecular representations for property prediction. His awards include the Presidential Early Career Award for Scientists and Engineers (2019), IJCAI Computers and Thought Award (2016), an NSF CAREER Award (2016), a Sloan Research Fellowship (2015), and a Microsoft Research Faculty Fellowship (2014). Zhang, Y., Liang, P., Chaudhuri, K., Sugiyama, M. On the Accuracy of Influence Functions for Measuring Group Effects. The worst form of professor. Certified Defenses for Data Poisoning Attacks. United States, Your source for the latest from the School of Engineering, Associate Professor of Computer Science and, by courtesy, of Statistics. 5 0 obj A., Haque, I. S., Beery, S., Leskovec, J., Kundaje, A., Pierson, E., Levine, S., Finn, C., Liang, P., Meila, M., Zhang, T. Beyond IID: Three Levels of Generalization for Question Answering on Knowledge Bases, Gu, Y., Kase, S., Vanni, M. T., Sadler, B. M., Liang, P., Yan, X., Su, Y., ACM, Prefix-Tuning: Optimizing Continuous Prompts for Generation, Li, X., Liang, P., Assoc Computat Linguist, Decoupling Exploration and Exploitation for Meta-Reinforcement Learning without Sacrifices. /Length 11 0 R His awards include the Presidential Early Career Award for Scientists and Engineers (2019), IJCAI Computers and Thought Award (2016), an NSF CAREER Award (2016), a Sloan Research Fellowship (2015), and a Microsoft Research Faculty Fellowship (2014). Mussmann, S., Liang, P., Bengio, S., Wallach, H., Larochelle, H., Grauman, K., CesaBianchi, N., Garnett, R. Semidefinite relaxations for certifying robustness to adversarial examples. III. Furthermore, we will review the use of iPSCs for development and testing of new therapeutic agents and the implications for high-throughput drug screening. A data structure for maintaining acyclicity in hypergraphs. Percy Liang is an Associate Professor of Computer Science and Statistics at Stanford University. /CreationDate (D:20230418051710-07'00') Liang, P., Bouchard-Ct, A., Klein, D., Taskar, B. His two research goals are (i) to make machine learning more robust, fair, and interpretable; and (ii) to make computers easier to communicate with through natural language. Stanford, CA 94305-4020Campus Map, Associate Professor, by courtesy, of Statistics, The Presidential Early Career Award for Scientists and Engineers (PECASE) embodies the high priority placed by the federal government on maintaining the leadership position of the United States in science by producing outstanding scientists and engineers and nurturing their continued developmen. Percy Liang: Stanford University Professor, technologist, and researcher in AI 7,897 views Mar 25, 2020 Stanford University Professor Percy Liang discusses the challenges of. Two students from his lab quit during their term because of his constant verbal abuse and harassment. Shi, T., Steinhardt, J., Liang, P., Lebanon, G., Vishwanathan, S. V. Environment-Driven Lexicon Induction for High-Level Instructions. His research seeks to develop trustworthy systems that can c. His awards include the Presidential Early Career Award for Scientists and Engineers (2019), IJCAI Computers and Thought Award (2016), an NSF CAREER Award (2016), a Sloan Research Fellowship (2015), and a Microsoft Research Faculty Fellowship (2014). View details for DOI 10.1161/CIRCRESAHA.112.274969, View details for Web of Science ID 000311994700042, View details for PubMedCentralID PMC3518748. He definetely is a pro! Previously, I received my B.S. A game-theoretic approach to generating spatial descriptions. Director, Center for Research on Foundation Models, Associate Professor of Computer Science, Stanford University. Liang, P., Tripp, O., Naik, M., Sagiv, M. Learning programs: a hierarchical Bayesian approach. Inferring Multidimensional Rates of Aging from Cross-Sectional Data. Understanding Self-Training for Gradual Domain Adaptation. Serafim Batzoglou. /N 3 << stream Davis, J., Gu, A., Choromanski, K., Dao, T., Re, C., Finn, C., Liang, P., Meila, M., Zhang, T. Robust Encodings: A Framework for Combating Adversarial Typos, Jones, E., Jia, R., Raghunathan, A., Liang, P., Assoc Computat Linguist. } 4(JR!$AkRf[(t Bw!hz#0 )l`/8p.7p|O~ Khani, F., Liang, P., Daume, H., Singh, A. Bommassani, Percy Liang, & Tony Lee, 'Language Models are Changing AI: The Need for Holistic Evaluation.' 12 OpenAI described weaponization risks of GPT-4 on p.12 of the "GPT-4 System Card." 13 See, e.g., the following benchmark for assessing adverse behaviors including power-seeking, disutility, and ethical violations: Wager, S., Fithian, W., Wang, S., Liang, P., Ghahramani, Z., Welling, M., Cortes, C., Lawrence, N. D., Weinberger, K. Q. PW Koh, S Sagawa, H Marklund, SM Xie, M Zhang, A Balsubramani, International Conference on Machine Learning, 5637-5664, Advances in neural information processing systems 30, E Choi, H He, M Iyyer, M Yatskar, W Yih, Y Choi, P Liang, L Zettlemoyer, Y Carmon, A Raghunathan, L Schmidt, JC Duchi, PS Liang, Advances in neural information processing systems 32, New articles related to this author's research, Squad: 100,000+ questions for machine comprehension of text, Understanding black-box predictions via influence functions, Know what you don't know: Unanswerable questions for SQuAD, Semantic parsing on freebase from question-answer pairs, Adversarial examples for evaluating reading comprehension systems, Prefix-tuning: Optimizing continuous prompts for generation, On the opportunities and risks of foundation models, Certified defenses against adversarial examples, Distributionally robust neural networks for group shifts: On the importance of regularization for worst-case generalization, Strategies for pre-training graph neural networks, Learning dependency-based compositional semantics, Dropout training as adaptive regularization, Wilds: A benchmark of in-the-wild distribution shifts, Certified defenses for data poisoning attacks, Unlabeled data improves adversarial robustness, Compositional semantic parsing on semi-structured tables, Delete, retrieve, generate: a simple approach to sentiment and style transfer. Wang, S., Wang, M., Wager, S., Liang, P., Manning, C. Video Event Understanding using Natural Language Descriptions. %PDF-1.4 A simple domain-independent probabilistic approach to generation. The sapogenins obtained from chlorogalum pomeridianum, Freeman Spogli Institute for International Studies, Institute for Computational and Mathematical Engineering (ICME), Institute for Human-Centered Artificial Intelligence (HAI), Institute for Stem Cell Biology and Regenerative Medicine, Stanford Institute for Economic Policy Research (SIEPR), Stanford Woods Institute for the Environment, Office of VP for University Human Resources, Office of Vice President for Business Affairs and Chief Financial Officer, Artificial Intelligence: Principles and Techniques, Writing Intensive Senior Research Project, Understanding and Developing Large Language Models, DOI 10.1146/annurev-linguist-030514-125312. ?_l) View details for DOI 10.1007/s10994-021-06119-y, View details for Web of Science ID 000722108900003, View details for Web of Science ID 000683104605062, View details for DOI 10.1145/3442381.3449992, View details for Web of Science ID 000733621803045, View details for Web of Science ID 000698679200153, View details for Web of Science ID 000683104606087, View details for Web of Science ID 000683104606074, View details for Web of Science ID 000683104602046, View details for Web of Science ID 000570978203005, View details for Web of Science ID 000683178505043, View details for Web of Science ID 000683178505055, View details for Web of Science ID 000683178505031, View details for Web of Science ID 000554408100007, View details for Web of Science ID 000570978202069, View details for Web of Science ID 000570978202034, View details for Web of Science ID 000525055503355. Percy Liang. The price of debiasing automatic metrics in natural language evaluation. rl1 Sharma, R., Gupta, S., Hariharan, B., Aiken, A., Liang, P., Nori, A. V. A data driven approach for algebraic loop invariants. We present a probabilistic model of diachronic phonology in which individual word forms undergo stochastic edits along the branches of a phylogenetic tree. A newly emerging application of iPSCs is in vitro disease modeling, which can significantly improve the never-ending search for new pharmacological cures. The following articles are merged in Scholar. Ramanathan, V., Joulin, A., Liang, P., Li Fei-Fei, F. F. Zero-shot Entity Extraction from Web Pages. On the UK Biobank human health dataset, our model reconstructs the observed data while learning interpretable rates of aging associated with diseases, mortality, and aging risk factors. My current research interests center around building a theory to understand and improve neural network models. /Producer (Apache FOP Version 1.0) I am associated with the Stanford Artificial Intelligence Lab and work with Tatsu Hashimoto and Percy Liang. from MIT, 2004; Ph.D. from UC Berkeley, 2011). 475 Via Ortega Data Recombination for Neural Semantic Parsing. Percy Liang honored with a Presidential Early Career Award. /Filter /FlateDecode Khani, F., Rinard, M., Liang, P., Erk, K., Smith, N. A. Wager, S., Fithian, W., Liang, P., Hazan, T., Papandreou, G., Tarlow, D. Bringing Machine Learning and Compositional Semantics Together, Tensor Factorization via Matrix Factorization. We prove that when this nonlinear function is constrained to be order-isomorphic, the model family is identifiable solely from cross-sectional data provided the distribution of time-independent variation is known. Pierson, E., Koh, P. W., Hashimoto, T., Koller, D., Leskovec, J., Eriksson, N., Liang, P. Kulal, S., Pasupat, P., Chandra, K., Lee, M., Padon, O., Aiken, A., Liang, P., Wallach, H., Larochelle, H., Beygelzimer, A., d'Alche-Buc, F., Fox, E., Garnett, R. NEURAL INFORMATION PROCESSING SYSTEMS (NIPS). Not sure what you can learn given his confusing behavior. Kuleshov, V., Chaganty, A., Liang, P., Lebanon, G., Vishwanathan, S. V. Learning Where to Sample in Structured Prediction. Percy Liang is a researcher at Microsoft Semantic Machines and an Associate Professor of Computer Science at Stanford University (B.S. from MIT, 2004; Ph.D. from UC Berkeley, 2011). A semantic parser converts these explanations into programmatic labeling functions that generate noisy labels for an arbitrary amount of unlabeled data, which is used to train a classifier. A probabilistic approach to language change. Hancock, B., Varma, P., Wang, S., Bringmann, M., Liang, P., Re, C., Gurevych, Miyao, Y. Analyzing the errors of unsupervised learning. How Much is 131 Million Dollars? Genome Editing of Human Embryonic Stem Cells and Induced Pluripotent Stem Cells With Zinc Finger Nucleases for Cellular Imaging. Associate Professor of Computer Science, Stanford University. 500 Liang, P., Narasimhan, M., Shilman, M., Viola, P. Methods and experiments with bounded tree-width Markov networks. Molecular imaging has proven to be a vital tool in the characterization of stem cell behavior in vivo. Frostig, R., Wang, S., Liang, P., Manning, C. D., Ghahramani, Z., Welling, M., Cortes, C., Lawrence, N. D., Weinberger, K. Q. He often fails to control his emotion when interacting with others. Learning Symmetric Collaborative Dialogue Agents with Dynamic Knowledge Graph Embeddings. Pasupat, P., Liang, P., Toutanova, K., Wu, H. Berant, J., Liang, P., Toutanova, K., Wu, H. Altitude Training: Strong Bounds for Single-Layer Dropout. Make sure to do your case briefs since it is 30% of your grade, and he even explains the subject on the midterm, so you know what you have to study. On the interaction between norm and dimensionality: multiple regimes in learning. However, the integration of reporter genes has typically relied on random integration, a method that is associated with unwanted insertional mutagenesis and positional effects on transgene expression.To address this barrier, we used genome editing with zinc finger nuclease (ZFN) technology to integrate reporter genes into a safe harbor gene locus (PPP1R12C, also known as AAVS1) in the genome of human embryonic stem cells and human induced pluripotent stem cells for molecular imaging.We used ZFN technology to integrate a construct containing monomeric red fluorescent protein, firefly luciferase, and herpes simplex virus thymidine kinase reporter genes driven by a constitutive ubiquitin promoter into a safe harbor locus for fluorescence imaging, bioluminescence imaging, and positron emission tomography imaging, respectively. Training accurate classifiers requires many labels, but each label provides only limited information (one bit for binary classification). His research spans theoretical machine learning to practical natural language processing; topics include semantic parsing, question answering, machine translation, online learning, method of moments, approximate inference, His two research goals are (i) to make machine learning more robust, fair, and interpretable; and (ii) to make computers easier to communicate with through natural language. His two research goals are (i) to make machine learning more robust, fair, and interpretable; and (ii) to make computers easier to communicate with through natural language. Difficult course materials do not necessarily help one to improve and grow. endobj Alexandre Bouchard-Ct, Percy Liang, Tom Griffiths, Dan Klein. Rate My Professors Enter your school to get started I'd like to look up a professor by name Join the RMP Family Love RMP? "FV %H"Hr 
