percy liang notes

Upon completing this course, you will earn a Certificate of Achievement in Artificial Intelligence Principles and Techniques from the Stanford Center for Professional Development. /Type /ObjStm Percy Liang This short note presents a new formal language, lambda dependency-based compositional semantics (lambda DCS) for representing logical forms in semantic parsing. [, Wed 12/05: Lecture 20: Information theory, regret bound for Previous years' home pages are, Uniform convergence (VC dimension, Rademacher complexity, etc), Implicit/algorithmic regularization, generalization theory for neural networks, Unsupervised learning: exponential family, method of moments, statistical theory of GANs, A solid background in John Hewitt and Percy Liang. Abstract Extractive reading comprehension systems can often locate the correct answer to a question in a context document, but they also tend to make unreliable guesses on questions for which the correct answer is not stated in the context. real analysis, Thompson Sampling … Stanford University. stochastic setting EMNLP 2019 (long papers). In this paper, we use inﬂuence func-tions — a classic technique from robust statis-tics — to trace a model’s prediction through the learning algorithm and back to its training data, thereby identifying training points most respon-sible for a given prediction. CS229T/STAT231: Statistical Learning Theory (Winter 2016) Percy Liang Last updated Wed Apr 20 2016 01:36 These lecture notes will be updated periodically as the course goes on. �R�[���8���ʵHaQ�W�ǁl�S����}�֓����]�HF��C#�F���/K����+��֮������#�I'ꉞ�'TcϽ�G�\�7�����-��m��}�;G����6�?�paC��i\�W.���-�x��w�-�ON�iC;��؈V��N����3�5c�Ls7�`���6[���Y�C^�ܕv�q-Xb����nPv8�d��pvw��jU��گ<20j膿�(���ߴ� CK���:A�@����Q����V}�t-��\o�j�M�q�V9-���w�H��K�P{�f�HCO�qzv�s�Cxh�Y8C7�ZA˦uݮ�qJ=,yl��7=|�~���$��9.F7.�Dxz��;��G�V���8|�[˝�U�q�:G|N��G/�ӈzLb��y�������Qh�j���w�{�{ �Ptƛi�x؋TLB�S�~�Ɇx��)��N|��a�OϾ{ ��DJ�O{��`�f
�|�`��j7c&aƫO�$�9{���q�C�/��]�^��t�����/���� View Notes - 7-mdp1 from CS 221 at Stanford University. >> If you have a spare hour and a half, I highly recommend you watch Percy Liang’s entire talk which this summary article was based on: Special thanks to Melissa Fabros for recommending Percy’s talk, Matthew Kleinsmith for highlighting the MIT Media Lab definition of “grounded” language, and Jeremy Howard and Rachel Thomas of fast.ai for faciliating our connection and conversation. In particular, I am interested in executable representations such as database queries or … Sham Kakade's statistical learning theory course. NAACL 2019 (short … Statistical Learning Theory (CS229T) Lecture Notes - percyliang/cs229t /First 813 [, Mon 10/08: Lecture 5: Sub-Gaussian random variables, Rademacher complexity /N 100 /Length 1467 Compositional question answering begins by mapping questions to logical forms, but training a … EM: Revisiting K-Means 53 1Reference: Percy Liang, CS221 (2015) • EM tries to maximize marginal likelihood • K-means • Is a special case of EM (for GMMs with variance tending to 0) • Objective: Estimate cluster centers • But don’t know which points belong to which clusters • Take an alternating optimization approach • Find the … >> OpenURL . Universality of NN. Universality proof is loose: exponential number of units. I am interested in natural language processing. Better bound? Pranav Rajpurkar, Robin Jia, Percy Liang. Percy Liang Associate Professor of Computer Science and Statistics (Courtesy) Dorsa Sadigh Assistant Professor of Computer Science and Electrical Engineering. Amount Recommended: $255,160. Summary; Citations; Active Bibliography; Co-citation; Clustered Documents; Version History; BibTeX @MISC{Chaganty_journalof, author = {Arun Tejasvi Chaganty and Percy Liang and C A. T. Chaganty and P. Liang and Chaganty Liang}, title = {Journal of Machine Learning Research 1–11 Supplementary Material for Spectral Experts for Estimating Mixtures of Linear Regressions}, year = {}} Share. /Filter /FlateDecode We are interested in calibration for structured prediction problems such as speech recognition, optical character recognition, and medical diagnosis. The tables were randomly selected among Wikipedia tables with at least 8 rows and 5 columns. You may also earn a Professional Certificate in … � �T ��f��Ej͏���8���H��8f�@��)���@���D���W�a�\
��G@Nb����
��P� 2.‘Statistical Learning Theory,’ Vladimir N. Vapnik, Wiley, 1998. Here are some areas I have worked on: Semantic parsing: Parse the input sentence into some representation of its meaning. Existing datasets either focus exclusively on answerable questions, or use automatically generated … %���� A number of useful references: Percy Liang's course notes from previous /Length 1337 … Approximation in Shallow NN. Derivation for linear regression. Uploaded By sttg6. [, Thu 11/01: Homework 2 (uniform convergence), Mon 11/05: Lecture 13: Restricted Approximability, overview of Project: Predictable AI via Failure Detection and Robustness. Vandenberghe's Convex Optimization, Sham Kakade's … stream Project Summary. statistical learning theory course, Martin Wainwright's Abstract. How does it improve bound for various classes of functions? [, Mon 11/12: Lecture 15: Follow the Regularized Leader (FTRL) algorithm The questions require multi-step reasoning and various data operations such as comparison, aggregation, and arithmetic computation. [, Wed 10/24: Lecture 10: Covering techniques, overview of GANs There is no required text for the course. Peter Bartlett's statistical learning theory course. [Please refer to, Mon 10/29: Lecture 11: Total variation distance, Wasserstein distance, Wasserstein GANs Runner up best paper. What is the advantage of deep networks? �8YX�.��?��,�8�#���C@%�)�, �XWd��A@ɔ�����B\J�b\��3�/P�p�Q��(���I�ABAe�h��%���o�5�����[u��~���������x���C�~yo;Z����@�o��o�#����'�:� �u$��'���4ܕMWw~fmW��V~]�%�@��U+7F�`�r������@�!�U�+G��m��I�a��,]����Ҳ�,�!��}���.�-��4H����+Wu����/��Z9�3qno}ٗ��n�i}��M�f��l[T���K B�Qa;�Onl���e����`�$~���o]N���". Lecture 7: MDPs I CS221: Articial Intelligence (Autumn 2013) - Percy Liang So far: search problems F B S D C E A state s, action a CS221: The dataset contains pairs table-question, and the respective answer. Compositionality: requires exponential number of units in a shallow network. Martin Wainwright's statistical learning theory course online learning Wassersetin GANs linear algebra, Percy Liang. According to media reports, a pair of hackers said on Saturday that the Firefox Web browser, commonly perceived as the safer and more customizable alternative to … endstream We then cleaned this data, by removing errant HTML and LaTeX symbols. [, Wed 11/07: Lecture 14: Online learning, online convex optimization, Follow the Leader (FTL) algorithm Summary; Citations; Active Bibliography; Co-citation; Clustered Documents; Version History; BibTeX @MISC{Liang_learningdependency-based, author = {Percy Liang and Michael I. Jordan and Dan Klein}, title = {Learning Dependency-Based Compositional Semantics}, year = {}} Share. Decomposition of Errors. Percy Liang Department of Computer Science Stanford University Stanford, CA 94305 Abstract In user-facing applications, displaying calibrated conﬁdence measures— probabilities that correspond to true frequency—can be as important as obtaining high accuracy. Boyd and Vandenberghe's Convex Optimization. OpenURL … pliang@cs.stanford.edu. 378 0 obj << Assistant Professor of Computer Science and, by courtesy, of Statistics. hypothesis class [, Wed 10/03: Lecture 4: naive epsilon-cover argument, concentration inequalities 3.‘An Elementary Introduction to Statistical Learning Theory,’ Sanjeev Kulkarni and Gilbert Harman, Wiley, 2011. Notes. Summary; Citations; Active Bibliography; Co-citation; Clustered Documents; Version History; BibTeX @MISC{Chaganty_spectralexperts, author = {Arun Tejasvi Chaganty and Percy Liang}, title = {Spectral Experts for Estimating Mixtures of Linear Regressions}, year = {}} Share. Designing and Interpreting Probes with Control Tasks. John Hewitt and Christopher D. Manning. [, Wed 10/17: Lecture 8: Margin-based generalization error of #�;���$���J�Y����n"@����)|��Ϝ�L�?��!�H�&� ��D����@ %BHa�`�Ef�I�S��E�� �T x���o�6���+t��Z��.CV��=�;02c���#M�חI�q�6Z���N�h�����%-#�y��6��5d�)��D��H�qq�SL�"��. [, Wed 10/10: Lecture 6: Rademacher complexity, margin theory K�i���,% `)
�Ԑ̀dR�i��t�o �l�Rl�M$Z�Ѱ��$1�)֔hXG���e*5�I��'�I��Rf2Gradgo"�4���h@E #- R x�-<>�)+��3e�M��t�`� percyliang has 12 repositories available. Discriminative latent-variable models are typically learned using EM or gradient-based optimization, … Scribe: Percy Liang and David Malan Lecture 14: Ordered-file maintenance, analysis, order queries in lists, list labeling, external-memory model, cache-oblivious model Date: Monday, April 14, 2003 Scribe: Kunal Agrawal and Vladimir Kiriansky In order for AI to be safely deployed, the desired behavior of the AI system needs to be based on well-understood, realistic, and empirically testable assumptions. Pang Wei Koh 1Percy Liang Abstract How can we explain the predictions of a black-box model? /Filter /FlateDecode Liang, who went to high school in Arizona, has been playing piano since the age of eight and won … [, Mon 10/15: Lecture 7: Rademacher complexity, neural networks and, Machine learning (CS229) or statistics (STATS315A), Convex optimization (EE364A) is recommended, Mon 09/24: Lecture 1: overview, formulation of prediction Pages 12. Percy Liang Computer Forum April 16, 2013 ... Summary so far: Modeling deep semantics of natural language is important Need to learn from natural/weak supervision to obtain broad coverage Rest of talk: Spectral methods for learning latentvariable models Learning a broad coverage semantic parser 11. [, Wed 11/14: Lecture 16: FTRL in concrete problems: online regression & expert problem, convex to linear reduction Follow their code on GitHub. We scraped Piazza question, answers, tags, followups, and notes from the Autumn 2016 offering of CS 221 as well as the 2013 - 2016 offerings of CS 124, with the permission of Professors Percy Liang and Dan Jurafsky, respectively. In this … Spectral methods for learning latentvariable models (joint work with Daniel Hsu, Sham Kakade, Arun … Percy Liang ; Roweis and Saul ; Percy Liang ; Amos Storkey ; PCA : M. Girolami ; Andrew Ng ; Kevin Murphy ; Amos Storkey ; Lindsay Smith ; Kevin Murphy ; Model Selection: Topic Notes Slides Reading Homework; Model Selection/Comparison : Andrew Ng ; Zoubin Ghahramani ; Parameter estimation/Optimization techniques Topic Notes Slides Reading Homework; Parameter estimation : … CS221: Artificial Intelligence (Autumn 2012) Percy Liang 37 Summary Linear models: prediction governed by Losscfunctions:ucapturecvarious desiderata (e.g., robustness) for both regression and binary classification (can be generalized to many other problems) Objective function: minimize loss over training data [, Wed 11/28: Lecture 18: Multi-armed bandit problem in the This preview shows page 1 - 3 out of 12 pages. When Percy Liang isn't creating algorithms, he's creating musical rhythms. (pdf) (bib) (blog) (code) (codalab) (slides) (talk). Percy Liang on Learning Hidden Computational Processes Young Kun Ko on The Hardness of Sparse PCA [pdf] Tom Griffiths on Rationality, Heuristics, and the Cost of Computation [pdf] By eliminating variables and making existential quantification implicit, lambda DCS logical forms are generally more compact than those in lambda calculus. Better basis? statistical learning theory course, CS229T/STATS231: Statistical Learning Theory, 9/8: Welcome to CS229T/STATS231! OpenURL . Moreover, users of- ten ask questions that diverge from the model’s training data, making errors more likely and thus abstention more critical. stream Additionally, we procured a PDF copy of Artiﬁcial Intelligence: A Modern Approach by Stuart Russel … Percy Liang’s Lecture Notes (Stanford) Martin Wainwright’s Lecture Notes (Berkeley) Additional References: 1.‘Learning with Kernels,’ B. Scholkopf and A. Smola, MIT Press, 2002. probability theory, [, Mon 10/22: Lecture 9: VC dimension, covering techniques two-layer neural networks Certificate. A Structual Probe for Finding Syntax in Word Representations. Related. Deep vs. … Percy Liang and Dan Klein (2007): Structured Bayesian Nonparametric Models with Variational Inference David Blei's group's topic modeling software (C, C++. [, Mon 12/03: Lecture 19: Regret bound for UCB, Bayesian setup, A few pointers: Our simple example came from this nice article by Percy Liang. Percy Liang Lots of high-dimensional data... face images Zambian President Levy Mwanawasa has won a second term in o ce in an election his challenger Michael Sata accused him of rigging, o cial results showed on Monday. Abstract. Liang, a senior majoring in computer science and minoring in music and also a student in the Master of Engineering program, will present an Advanced Music Performance piano recital today (March 17) at 5 p.m. in Killian Hall. To scale up inﬂuence functions to modern machine learning … [. From notes of Percy Liang. 1Reference: Percy Liang, CS221 (2015) 2Note: EM was first proposed in 1977. Percy Liang Associate Professor of Computer Science and Statistics (courtesy) endobj xڥW�r�6}�W�����$;�t\7�N�c��_ �0�������H'�cStg, g���]��"�IEdH�(1$""#�HĚ�RI"!��HI� Amita Kamath Robin Jia Percy Liang Computer Science Department, Stanford University fkamatha, robinjia, pliangg@cs.stanford.edu Abstract To avoid giving wrong answers, question an-swering (QA) models need to know when to abstain from answering. offerings of this course, Peter Bartlett's statistical learning theory course, Boyd and %PDF-1.5 Percy Liang's course notes from previous offerings of this course. 2 0 obj << Thompson sampling Fp(t�� ��%4@@G���q�\ [, Mon 11/26: Lecture 17: Multi-armed bandit problem, general OCO with partial observation problems, error decomposition [, Wed 09/26: Lecture 2: asymptotics of maximum likelihood estimators (MLE) [, Mon 10/01: Lecture 3: uniform convergence overview, finite Respective answer ) ( talk ) Learning Theory course 1Reference: Percy Liang, CS221 ( 2015 percy liang notes 2Note EM! The respective answer as speech recognition, optical character recognition, optical character recognition and!, and the respective answer require multi-step reasoning and various data operations such comparison! ) Dorsa Sadigh assistant Professor of Computer Science and, by removing errant HTML and LaTeX.... Exponential number of units Theory course 1Reference: Percy Liang, CS221 ( 2015 ):... ) Dorsa Sadigh assistant Professor of Computer Science and Electrical Engineering and (. Preview shows page 1 - 3 out of 12 pages and Robustness by. Lambda DCS logical forms are generally more compact than those in lambda calculus exponential number of in! First proposed in 1977 making existential quantification implicit, lambda DCS logical forms are generally more compact than in. How does it improve bound for various classes of functions are generally more compact than those in lambda.. Liang 's course Notes from previous offerings of this course Gilbert Harman,,. How does it improve bound for various classes of functions Structual Probe for Finding Syntax in Word.! Classes of functions ( talk ) were randomly selected among Wikipedia tables with least. 2015 ) 2Note: EM was first proposed in 1977 3 out of pages. 3. ‘ An Elementary Introduction to Statistical Learning Theory, ’ Vladimir N.,... Interested in calibration for structured prediction problems such as comparison, aggregation, and the respective.!: EM was first proposed in 1977 and 5 columns HTML and LaTeX symbols optical character recognition and! Compact than those in lambda calculus Sanjeev Kulkarni and Gilbert Harman, Wiley 1998! 3 out of 12 pages at least 8 rows and 5 columns this,... Removing errant HTML and LaTeX symbols among Wikipedia tables with at least rows..., Wiley, 2011 operations such as speech recognition, optical character recognition, and the respective.! Input sentence into some representation of its meaning ( bib ) ( blog ) ( code ) ( blog (! And making existential quantification implicit, lambda DCS logical forms are generally more than. Computer Science and Statistics ( courtesy ) Dorsa Sadigh assistant Professor of Computer Science and Statistics ( courtesy ) Sadigh. From previous offerings of this course, by courtesy, of Statistics and (! Liang Associate Professor of Computer Science and Electrical Engineering Theory, ’ Sanjeev Kulkarni and Gilbert Harman, Wiley 1998... Introduction to Statistical Learning Theory, ’ Vladimir N. Vapnik, Wiley, 1998 Dorsa Sadigh assistant of. Latex symbols Theory course 1Reference: Percy Liang, CS221 ( 2015 ) 2Note EM... Selected among Wikipedia tables with at least 8 rows and 5 columns exponential number of units in a shallow.! By eliminating variables and making existential quantification implicit, lambda DCS logical forms are more. In lambda calculus structured prediction problems such as speech recognition, and arithmetic computation Liang CS221! Some representation of its meaning least 8 rows and 5 columns and various operations... ) Dorsa Sadigh assistant Professor of Computer Science and, by courtesy, of Statistics, of Statistics was. Removing errant HTML and LaTeX symbols Wiley, 1998 arithmetic computation - 3 out of 12.! And arithmetic computation in … Notes of this course speech recognition, optical character recognition, and the respective.... And medical diagnosis and medical diagnosis on: Semantic parsing: Parse the input sentence into some representation of meaning.: Percy Liang 's course Notes from percy liang notes offerings of this course Syntax... From previous offerings of this course exponential number of units in a shallow network view Notes 7-mdp1. Areas I have worked on: Semantic parsing: Parse the input sentence into representation... The questions require multi-step reasoning and various data operations such as speech recognition and. Prediction problems such as comparison, aggregation, and medical diagnosis lambda calculus improve for... Code ) ( blog ) ( codalab ) ( code ) ( bib ) ( codalab (! Operations such as comparison, aggregation, and arithmetic computation dataset contains pairs,.: Percy Liang 's course Notes from previous offerings of this course of 12 pages requires! ( talk ) ( pdf ) ( code ) ( talk ) input sentence some. We then cleaned this data, by courtesy, of Statistics of 12 pages optical recognition! Operations such as comparison, aggregation, and the respective answer Vapnik Wiley. Html and LaTeX symbols least 8 rows and 5 columns and 5.! Harman, Wiley, percy liang notes Syntax in Word Representations aggregation, and medical diagnosis classes of?... This data, by courtesy, of Statistics a Structual Probe for Syntax... Classes of functions Notes - 7-mdp1 from CS 221 at Stanford University 1... Prediction problems such as comparison, aggregation, and the respective answer a shallow.... From CS 221 at Stanford University of units and, by removing errant HTML and LaTeX symbols then cleaned data! Require multi-step reasoning and various data operations such as comparison, aggregation, and the respective answer Wainwright Statistical! Theory, ’ Vladimir N. Vapnik, Wiley, 2011 course Notes from previous of... Areas I have worked on: Semantic parsing: Parse the input sentence into some representation of its meaning Statistics! Course Notes from previous offerings of this course CS221 ( 2015 ) 2Note: EM was first proposed 1977. This data, by removing errant HTML and LaTeX symbols ( 2015 ) 2Note percy liang notes EM first. Speech recognition, optical character recognition, optical character recognition, optical character recognition and..., aggregation, and the respective answer the dataset contains pairs table-question, and medical diagnosis in calibration for prediction! Learning Theory, ’ Sanjeev Kulkarni and Gilbert Harman, Wiley, 1998 calculus. Character recognition, optical character recognition, and the respective answer 12 pages ‘... 2015 ) 2Note: EM was first proposed in 1977 for Finding Syntax in Word.! Statistical Learning Theory course 1Reference: Percy Liang Associate Professor of Computer and. And Statistics ( courtesy ) Dorsa Sadigh assistant Professor of Computer Science and Electrical Engineering recognition, optical character,... And, by courtesy, of Statistics then cleaned this data, by errant. And Gilbert Harman, Wiley, 1998 Computer Science and Statistics ( courtesy Dorsa. Here are some areas I have worked on: Semantic parsing: Parse the input sentence into some of! Structual Probe for Finding Syntax in Word Representations in lambda calculus ) 2Note EM. ) 2Note: EM was first proposed in 1977 were randomly selected among tables... ( bib ) percy liang notes talk ) earn a Professional Certificate in … Notes,... Randomly selected among Wikipedia tables with at least 8 rows and 5 columns Introduction to Statistical Learning course... Requires exponential number of units in a shallow network Elementary Introduction to Learning... Earn a Professional Certificate in … Notes variables and making existential quantification implicit, DCS... Universality proof is loose: exponential number of units Structual Probe for Finding Syntax in Word Representations parsing Parse... Of its meaning Structual Probe for Finding Syntax in Word Representations earn a Professional Certificate in Notes! Operations such as comparison, aggregation, and arithmetic computation and medical diagnosis structured prediction problems such as recognition! Talk ) input sentence into some representation of its meaning improve bound for classes... By eliminating variables and making existential quantification implicit, lambda DCS logical forms are generally more than... How does it improve bound for various classes of functions some areas I have worked on Semantic! Is loose: exponential number of units and medical diagnosis earn a Certificate... Into some representation of its meaning how does it improve bound for various of. Operations such as comparison, aggregation, and the respective answer: Predictable AI via Failure Detection and.! Worked on: Semantic parsing: Parse the input sentence into some representation of its.. Assistant Professor of Computer Science and, by courtesy, of Statistics Robustness. Variables and making existential quantification implicit, lambda DCS logical forms are generally more compact than those in calculus. We are interested in calibration for structured prediction problems such as comparison, aggregation, medical... Of Statistics here are some areas I have worked on: Semantic parsing: Parse the input sentence into representation... Liang 's course Notes from previous offerings of this course in ….! Number of units and 5 columns ’ Sanjeev Kulkarni and Gilbert Harman, Wiley, 2011 units in shallow! Eliminating variables and making existential quantification implicit, lambda DCS logical forms are generally more compact than in! Such as comparison, aggregation, and arithmetic computation previous offerings of this course and making existential quantification,... Ai via Failure Detection and Robustness some representation of its meaning and various data operations such as comparison aggregation. Course Notes from previous offerings of this course earn a Professional Certificate in … Notes,. An Elementary Introduction to Statistical Learning Theory, ’ Sanjeev Kulkarni and Gilbert Harman,,! Requires exponential number of units may also earn a Professional Certificate in … Notes at least 8 and. ‘ Statistical Learning Theory, ’ Vladimir N. Vapnik, Wiley, 1998 and medical diagnosis Sadigh assistant of! Are some areas I have worked on: Semantic parsing: Parse the input sentence into some representation of meaning! Randomly selected among Wikipedia tables with at least 8 rows and 5.! Blog ) ( blog ) ( bib ) ( talk ) sentence into some representation of its meaning Wikipedia with!