Dynamic Poisson Factorization
|
|
- Britney Bryant
- 5 years ago
- Views:
Transcription
1 Dynamic Poisson Factorization Laurent Charlin Joint work with: Rajesh Ranganath, James McInerney, David M. Blei McGill & Columbia University Presented at RecSys 2015
2 2
3 Click Data for a paper 3.5 Item 4663: The Google Similarity Distance Access frequency The Google Similarity effect arxiv: cs/ Year
4 arxiv evolves over time Click data evolves over time Popularity of papers change over time Users interests change over time The same is also true of other datasets 4
5 Recommendations should evolve over time
6 3.5 Raw Data Item 4663: The Google Similarity Distance Modeled Data (e.g., Matrix Factorization) 0.14 Access frequency The Google Similarity effect arxiv: cs/ DDtD 6tructures GrDSh Theory ComSresseG 6eQsiQg 4uDQtum CryStogrDShy
7 3.5 Raw Data Item 4663: The Google Similarity Distance Modeled Data (e.g., Matrix Factorization) 0.14 Access frequency The Google Similarity effect arxiv: cs/ DDtD 6tructures GrDSh Theory ComSresseG 6eQsiQg 4uDQtum CryStogrDShy Time-aware Modeled Data Factor expression
8 Objectives Understand click data over time Better predictions of future user interests Explore how (large) datasets evolve over time Motivated by the arxiv Study other datasets 7
9 Technical approach Dynamic model of user preferences Poisson matrix factorization [Gopalan et al. UAI 15] User and item latent factors evolve over time Develop inference procedure Fit this model using data Explore fits 8
10 Data (e.g., clicks) Recommendations Generative Model a, b u y M Learning & Inference Exploration (latent factors) c, d v N Factor expression
11 Background 10
12 Gaussian Matrix Factorization [Salakhutdinov et al. 15] a, b u y M c, d v N [wikipedia] 11
13 Gaussian Matrix Factorization [Salakhutdinov et al. 15] a, b u Poisson Matrix Factorization (PF) [Gopalan et al. 15] y M c, d v N [wikipedia] 11 [wikipedia]
14 Properties of PF Suitable for implicit and explicit data Model can be fit to massive datasets 12
15 9 User Click frequency
16 Dynamic Poisson factorization 14
17 Users preferences evolve over time Item popularity evolves over time Ratings are not exchangeable over time 15
18 Dynamic Matrix Factorization (DPF) 16
19 Dynamic Matrix Factorization (DPF) y 0 y 1... y T 16
20 Dynamic Matrix Factorization (DPF) u 0 u 1... u T y 0 y 1... y T M v 0 v 1... v T N 16
21 Dynamic Matrix Factorization (DPF) a, b u 0 u 1... u T y 0 y 1... y T M c, d v 0 v 1... v T N Modeling: Smooth evolution through time 16
22 Smooth Evolution through time Gaussian time series model Exponentiate to obtain positive Poisson rates 17
23 Local and Global factors Two levels of modeling: Local factors: model evolution through time Global factors: model constant preferences wrt time µū, ū ū µ u, u u 0 u 1... u T y 0 y 1... y T M µ v, v v 0 v 1... v T v N µ v, v 18
24 Local and Global factors Two levels of modeling: Local factors: model evolution through time Global factors: model constant preferences wrt time µū, ū ū µ u, u u 0 u 1... u T y 0 y 1... y T M µ v, v v 0 v 1... v T v N µ v, v 18
25 Local and Global factors Two levels of modeling: Local factors: model evolution through time Global factors: model constant preferences wrt time µū, ū ū µ u, u u 0 u 1... u T y 0 y 1... y T M µ v, v v 0 v 1... v T v N µ v, v 18
26 A 3 timeslice Dynamic PF µū, ū ū µ u, u u 0 u 1 u 2 y 0 y 1 y 2 M µ v, v v 0 v 1 v 2 v N µ v, v
27 Inference Variational inference algorithm Mean-field (factorized) approximation Non-conjugacy between and Use an additional approximation See the paper for details on the algorithm Scales linearly in time, we provide a parallel implementation 20
28 Empirical results 21
29 Explicit data Augmented matrix factorization [Koren 10, Li 11, Gultekin 14] Tensor factorization [Karatzoglou 10] Bayesian Tensor Probabilistic Factorization [Xiong 10] Implicit data HMM [Sahoo 12] Gamma process MF [Acharya 15] DPF (us): User and item dynamicity Implicit and explicit Scalable 22
30 Experimental setup Test on multiple time steps Train data: clicks up to time slice t Predict held-out clicks at time t+1 Metrics for implicit data Ranking-based (NDCG, Recall, MAR, MRR) 23
31 Netflix-time MAR MRR NDCG dpf BPTF [Xiong 10] PF-all [Gopalan 15] PF-last [Gopalan 15] K users, 3.6K items, 2M non-zero ratings, binarized 3 months per time slice 24
32 Netflix MAR MRR NDCG dpf PF-all [Gopalan 15] PF-last [Gopalan 15] K users, 14K items, 6.9M non-zero ratings, binarized 6 months per time slice 25
33 arxiv MAR MRR NDCG dpf PF-all [Gopalan 15] PF-last [Gopalan 15] K users, 5K items, 1.3M clicks 18 months per time slice 26
34 Usage Data DPF Factors 3.5 The Google Similarity effect arxiv: cs/ Item 4663: The Google Similarity Distance Factor expression 3.0 Access frequency GOobDO fdctor exsressioq DPF Global Factors GrDSh Theory 4uDQtuP & &ops. &opsoexity DDtD 6tructures & ProbDbiOity GeQerDO 0Dth 2StiPizDtioQ
35 Paper Selected at Random FaFtor expression
36 FaFtor expression
37 FaFtor expression he entrrpy frrmuld frr the 5icci flrw Dnd its germetric DpplicDtiRns (mdth/ ) 6 5 Access fretuency DDte
38 FaFtor expression he entrrpy frrmuld frr the 5icci flrw Dnd its germetric DpplicDtiRns (mdth/ ) 6 5 Access fretuency DDte
39 FaFtor expression Perelman 6 declines the 5 Fields Medal Access fretuency 4 3 7he entrrpy frrmuld frr the 5icci flrw Dnd its germetric DpplicDtiRns (mdth/ ) DDte
40 FaFtor expression Perelman 6 declines the 5 Fields Medal Access fretuency 4 3 7he entrrpy frrmuld frr the 5icci flrw Dnd its germetric DpplicDtiRns (mdth/ ) Perelman rejects the Millenium Prize DDte
41 Usage Data DPF Factors 9 User Click frequency Factor expression
42 Fields evolve over time Computability Classes Physics & Society Quantum Information Theory Quantum Logic Quantum Algorithms General Interest CS Graph Theory Information Theory Message Passing Algorithms Mathematical Statistics Information Theory Computational Geometry Probability & Statistics for Graphs Monte Carlo & Simulation Community Detection Quantum Computing Machine Learning & Information Theory Computational Complexity Condensed Matter & Computing Network Analysis 04/05 10/05 04/06 10/06 04/07 10/07 04/08 10/08 04/09 10/09 04/10 10/10 04/11 10/11 03/12 31
43 Fields evolve over time Computability Classes Physics & Society Quantum Information Theory Quantum Logic Quantum Algorithms General Interest CS Graph Theory Information Theory Message Passing Algorithms Mathematical Statistics Information Theory Computational Geometry Probability & Statistics for Graphs Monte Carlo & Simulation Community Detection Quantum Computing Machine Learning & Information Theory Computational Complexity Condensed Matter & Computing Network Analysis 04/05 10/05 04/06 10/06 04/07 10/07 04/08 10/08 04/09 10/09 04/10 10/10 04/11 10/11 03/12 31
44 Conclusions Dynamic Poisson factorization (DPF) 9 User Click frequency Factor expression Implicit data, scalable Open source implementation is available: github.com/blei-lab/ Future work: levels of granularities, continuous time 32
45 Thank you! Dynamic Poisson factorization (DPF) 9 User Click frequency Factor expression Implicit data, scalable Open source implementation is available: github.com/blei-lab/ Future work: levels of granularities, continuous time 32
Content-based Recommendation
Content-based Recommendation Suthee Chaidaroon June 13, 2016 Contents 1 Introduction 1 1.1 Matrix Factorization......................... 2 2 slda 2 2.1 Model................................. 3 3 flda 3
More informationAn Overview of Edward: A Probabilistic Programming System. Dustin Tran Columbia University
An Overview of Edward: A Probabilistic Programming System Dustin Tran Columbia University Alp Kucukelbir Eugene Brevdo Andrew Gelman Adji Dieng Maja Rudolph David Blei Dawen Liang Matt Hoffman Kevin Murphy
More information* Matrix Factorization and Recommendation Systems
Matrix Factorization and Recommendation Systems Originally presented at HLF Workshop on Matrix Factorization with Loren Anderson (University of Minnesota Twin Cities) on 25 th September, 2017 15 th March,
More informationScaling Neighbourhood Methods
Quick Recap Scaling Neighbourhood Methods Collaborative Filtering m = #items n = #users Complexity : m * m * n Comparative Scale of Signals ~50 M users ~25 M items Explicit Ratings ~ O(1M) (1 per billion)
More informationMixed Membership Matrix Factorization
Mixed Membership Matrix Factorization Lester Mackey 1 David Weiss 2 Michael I. Jordan 1 1 University of California, Berkeley 2 University of Pennsylvania International Conference on Machine Learning, 2010
More informationRecurrent Latent Variable Networks for Session-Based Recommendation
Recurrent Latent Variable Networks for Session-Based Recommendation Panayiotis Christodoulou Cyprus University of Technology paa.christodoulou@edu.cut.ac.cy 27/8/2017 Panayiotis Christodoulou (C.U.T.)
More informationMixed Membership Matrix Factorization
Mixed Membership Matrix Factorization Lester Mackey University of California, Berkeley Collaborators: David Weiss, University of Pennsylvania Michael I. Jordan, University of California, Berkeley 2011
More informationLarge-scale Ordinal Collaborative Filtering
Large-scale Ordinal Collaborative Filtering Ulrich Paquet, Blaise Thomson, and Ole Winther Microsoft Research Cambridge, University of Cambridge, Technical University of Denmark ulripa@microsoft.com,brmt2@cam.ac.uk,owi@imm.dtu.dk
More informationA Gradient-based Adaptive Learning Framework for Efficient Personal Recommendation
A Gradient-based Adaptive Learning Framework for Efficient Personal Recommendation Yue Ning 1 Yue Shi 2 Liangjie Hong 2 Huzefa Rangwala 3 Naren Ramakrishnan 1 1 Virginia Tech 2 Yahoo Research. Yue Shi
More informationSQL-Rank: A Listwise Approach to Collaborative Ranking
SQL-Rank: A Listwise Approach to Collaborative Ranking Liwei Wu Depts of Statistics and Computer Science UC Davis ICML 18, Stockholm, Sweden July 10-15, 2017 Joint work with Cho-Jui Hsieh and James Sharpnack
More informationRestricted Boltzmann Machines for Collaborative Filtering
Restricted Boltzmann Machines for Collaborative Filtering Authors: Ruslan Salakhutdinov Andriy Mnih Geoffrey Hinton Benjamin Schwehn Presentation by: Ioan Stanculescu 1 Overview The Netflix prize problem
More informationScalable Bayesian Matrix Factorization
Scalable Bayesian Matrix Factorization Avijit Saha,1, Rishabh Misra,2, and Balaraman Ravindran 1 1 Department of CSE, Indian Institute of Technology Madras, India {avijit, ravi}@cse.iitm.ac.in 2 Department
More informationCausal Inference and Recommendation Systems. David M. Blei Departments of Computer Science and Statistics Data Science Institute Columbia University
Causal Inference and Recommendation Systems David M. Blei Departments of Computer Science and Statistics Data Science Institute Columbia University EF In: Ratings or click data Out: A system that provides
More informationarxiv: v2 [cs.ir] 14 May 2018
A Probabilistic Model for the Cold-Start Problem in Rating Prediction using Click Data ThaiBinh Nguyen 1 and Atsuhiro Takasu 1, 1 Department of Informatics, SOKENDAI (The Graduate University for Advanced
More informationDecoupled Collaborative Ranking
Decoupled Collaborative Ranking Jun Hu, Ping Li April 24, 2017 Jun Hu, Ping Li WWW2017 April 24, 2017 1 / 36 Recommender Systems Recommendation system is an information filtering technique, which provides
More informationProbabilistic Matrix Factorization
Probabilistic Matrix Factorization David M. Blei Columbia University November 25, 2015 1 Dyadic data One important type of modern data is dyadic data. Dyadic data are measurements on pairs. The idea is
More informationModeling User Exposure in Recommendation
Modeling User Exposure in Recommendation ABSTRACT Dawen Liang Columbia University New York, NY dliang@ee.columbia.edu James McInerney Columbia University New York, NY james@cs.columbia.edu Collaborative
More informationCollaborative topic models: motivations cont
Collaborative topic models: motivations cont Two topics: machine learning social network analysis Two people: " boy Two articles: article A! girl article B Preferences: The boy likes A and B --- no problem.
More informationPractical Bayesian Optimization of Machine Learning. Learning Algorithms
Practical Bayesian Optimization of Machine Learning Algorithms CS 294 University of California, Berkeley Tuesday, April 20, 2016 Motivation Machine Learning Algorithms (MLA s) have hyperparameters that
More information13: Variational inference II
10-708: Probabilistic Graphical Models, Spring 2015 13: Variational inference II Lecturer: Eric P. Xing Scribes: Ronghuo Zheng, Zhiting Hu, Yuntian Deng 1 Introduction We started to talk about variational
More informationSTA 4273H: Statistical Machine Learning
STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Computer Science! Department of Statistical Sciences! rsalakhu@cs.toronto.edu! h0p://www.cs.utoronto.ca/~rsalakhu/ Lecture 7 Approximate
More informationBayesian Learning and Inference in Recurrent Switching Linear Dynamical Systems
Bayesian Learning and Inference in Recurrent Switching Linear Dynamical Systems Scott W. Linderman Matthew J. Johnson Andrew C. Miller Columbia University Harvard and Google Brain Harvard University Ryan
More informationMatrix Factorization Techniques for Recommender Systems
Matrix Factorization Techniques for Recommender Systems By Yehuda Koren Robert Bell Chris Volinsky Presented by Peng Xu Supervised by Prof. Michel Desmarais 1 Contents 1. Introduction 4. A Basic Matrix
More informationRecommender Systems EE448, Big Data Mining, Lecture 10. Weinan Zhang Shanghai Jiao Tong University
2018 EE448, Big Data Mining, Lecture 10 Recommender Systems Weinan Zhang Shanghai Jiao Tong University http://wnzhang.net http://wnzhang.net/teaching/ee448/index.html Content of This Course Overview of
More informationMatrix Factorization Techniques For Recommender Systems. Collaborative Filtering
Matrix Factorization Techniques For Recommender Systems Collaborative Filtering Markus Freitag, Jan-Felix Schwarz 28 April 2011 Agenda 2 1. Paper Backgrounds 2. Latent Factor Models 3. Overfitting & Regularization
More informationTime-aware Collaborative Topic Regression: Towards Higher Relevance in Textual Item Recommendation
Time-aware Collaborative Topic Regression: Towards Higher Relevance in Textual Item Recommendation Anas Alzogbi Department of Computer Science, University of Freiburg 79110 Freiburg, Germany alzoghba@informatik.uni-freiburg.de
More informationCollaborative Filtering. Radek Pelánek
Collaborative Filtering Radek Pelánek 2017 Notes on Lecture the most technical lecture of the course includes some scary looking math, but typically with intuitive interpretation use of standard machine
More informationA Randomized Approach for Crowdsourcing in the Presence of Multiple Views
A Randomized Approach for Crowdsourcing in the Presence of Multiple Views Presenter: Yao Zhou joint work with: Jingrui He - 1 - Roadmap Motivation Proposed framework: M2VW Experimental results Conclusion
More informationReview: Probabilistic Matrix Factorization. Probabilistic Matrix Factorization (PMF)
Case Study 4: Collaborative Filtering Review: Probabilistic Matrix Factorization Machine Learning for Big Data CSE547/STAT548, University of Washington Emily Fox February 2 th, 214 Emily Fox 214 1 Probabilistic
More informationSparse Stochastic Inference for Latent Dirichlet Allocation
Sparse Stochastic Inference for Latent Dirichlet Allocation David Mimno 1, Matthew D. Hoffman 2, David M. Blei 1 1 Dept. of Computer Science, Princeton U. 2 Dept. of Statistics, Columbia U. Presentation
More informationarxiv: v2 [stat.ml] 16 Aug 2016
The Bayesian Low-Rank Determinantal Point Process Mixture Model Mike Gartrell Microsoft mike.gartrell@acm.org Ulrich Paquet Microsoft ulripa@microsoft.com Noam Koenigstein Microsoft noamko@microsoft.com
More informationGenerative Models for Discrete Data
Generative Models for Discrete Data ddebarr@uw.edu 2016-04-21 Agenda Bayesian Concept Learning Beta-Binomial Model Dirichlet-Multinomial Model Naïve Bayes Classifiers Bayesian Concept Learning Numbers
More informationarxiv: v2 [stat.ml] 4 Feb 2016
Modeling User Exposure in Recommendation arxiv:1510.07025v2 [stat.ml] 4 Feb 2016 ABSTRACT Dawen Liang Columbia University New York, NY dliang@ee.columbia.edu James McInerney Columbia University New York,
More informationClustering based tensor decomposition
Clustering based tensor decomposition Huan He huan.he@emory.edu Shihua Wang shihua.wang@emory.edu Emory University November 29, 2017 (Huan)(Shihua) (Emory University) Clustering based tensor decomposition
More informationPATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 13: SEQUENTIAL DATA
PATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 13: SEQUENTIAL DATA Contents in latter part Linear Dynamical Systems What is different from HMM? Kalman filter Its strength and limitation Particle Filter
More informationContext-aware factorization methods for implicit feedback based recommendation problems
Context-aware factorization methods for implicit feedback based recommendation problems PhD thesis Balázs Hidasi Under the supervision of Dr. Gábor Magyar External advisor: Dr. Domonkos Tikk Budapest University
More informationTFMAP: Optimizing MAP for Top-N Context-aware Recommendation
TFMAP: Optimizing MAP for Top-N Context-aware Recommendation Yue Shi a, Alexandros Karatzoglou b, Linas Baltrunas b, Martha Larson a, Alan Hanjalic a, Nuria Oliver b a Delft University of Technology, Netherlands;
More informationCollaborative Filtering on Ordinal User Feedback
Proceedings of the Twenty-Third International Joint Conference on Artificial Intelligence Collaborative Filtering on Ordinal User Feedback Yehuda Koren Google yehudako@gmail.com Joseph Sill Analytics Consultant
More informationBayesian Nonparametric Poisson Factorization for Recommendation Systems
Bayesian Nonparametric Poisson Factorization for Recommendation Systems Prem Gopalan Francisco J. R. Ruiz Rajesh Ranganath David M. Blei Princeton University University Carlos III in Madrid Princeton University
More informationAndriy Mnih and Ruslan Salakhutdinov
MATRIX FACTORIZATION METHODS FOR COLLABORATIVE FILTERING Andriy Mnih and Ruslan Salakhutdinov University of Toronto, Machine Learning Group 1 What is collaborative filtering? The goal of collaborative
More informationBayesian Networks BY: MOHAMAD ALSABBAGH
Bayesian Networks BY: MOHAMAD ALSABBAGH Outlines Introduction Bayes Rule Bayesian Networks (BN) Representation Size of a Bayesian Network Inference via BN BN Learning Dynamic BN Introduction Conditional
More informationLocally Private Bayesian Inference for Count Models
Locally Private Bayesian Inference for Count Models Aaron Schein UMass Amherst Zhiwei Steven Wu Microsoft Research Mingyuan Zhou UT Austin Hanna Wallach Microsoft Research 1 Introduction Data from social
More informationVCMC: Variational Consensus Monte Carlo
VCMC: Variational Consensus Monte Carlo Maxim Rabinovich, Elaine Angelino, Michael I. Jordan Berkeley Vision and Learning Center September 22, 2015 probabilistic models! sky fog bridge water grass object
More informationApproximate Bayesian inference
Approximate Bayesian inference Variational and Monte Carlo methods Christian A. Naesseth 1 Exchange rate data 0 20 40 60 80 100 120 Month Image data 2 1 Bayesian inference 2 Variational inference 3 Stochastic
More informationFactor Modeling for Advertisement Targeting
Ye Chen 1, Michael Kapralov 2, Dmitry Pavlov 3, John F. Canny 4 1 ebay Inc, 2 Stanford University, 3 Yandex Labs, 4 UC Berkeley NIPS-2009 Presented by Miao Liu May 27, 2010 Introduction GaP model Sponsored
More informationProbabilistic Canonical Tensor Decomposition for Predicting User Preference
Probabilistic Canonical Tensor Decomposition for Predicting User Preference Maja Rita Rudolph mrr2163@columbia.edu John Paisley jpaisley@columbia.edu San Gultekin sg3108@columbia.edu Shih-Fu Chang sfchang@columbia.edu
More informationBayesian nonparametric models for bipartite graphs
Bayesian nonparametric models for bipartite graphs François Caron Department of Statistics, Oxford Statistics Colloquium, Harvard University November 11, 2013 F. Caron 1 / 27 Bipartite networks Readers/Customers
More informationA Bayesian Latent Variable Model of User Preferences with Item Context
A Bayesian Latent Variable Model of User Preferences with Item Context Aghiles Salah and Hady W. Lauw School of Information Systems, Singapore Management University, Singapore {asalah, hadywlauw}@smu.edu.sg
More informationMatrix and Tensor Factorization from a Machine Learning Perspective
Matrix and Tensor Factorization from a Machine Learning Perspective Christoph Freudenthaler Information Systems and Machine Learning Lab, University of Hildesheim Research Seminar, Vienna University of
More informationSTA 4273H: Statistical Machine Learning
STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Statistics! rsalakhu@utstat.toronto.edu! http://www.utstat.utoronto.ca/~rsalakhu/ Sidney Smith Hall, Room 6002 Lecture 7 Approximate
More informationScalable Recommendation with Hierarchical Poisson Factorization
Scalable Recommendation with Hierarchical Poisson Factorization Prem Gopalan Department of Computer Science Princeton University Princeton, NJ Jake M. Hofman Microsoft Research 641 Sixth Avenue, Floor
More informationWhy Aren t You Using Probabilistic Programming? Dustin Tran Columbia University
Why Aren t You Using Probabilistic Programming? Dustin Tran Columbia University Alp Kucukelbir Adji Dieng Dave Moore Dawen Liang Eugene Brevdo Ian Langmore Josh Dillon Maja Rudolph Brian Patton Srinivas
More informationRecurrent Coevolutionary Latent Feature Processes for Continuous-Time Recommendation
Recurrent Coevolutionary Latent Feature Processes for Continuous-Time Recommendation Hanjun Dai, Yichen Wang, Rashit Trivedi, Le Song College of Computing, Georgia Institute of Technology {yichen.wang,
More information20: Gaussian Processes
10-708: Probabilistic Graphical Models 10-708, Spring 2016 20: Gaussian Processes Lecturer: Andrew Gordon Wilson Scribes: Sai Ganesh Bandiatmakuri 1 Discussion about ML Here we discuss an introduction
More informationGaussian Process Factorization Machines for Context-aware Recommendations
Gaussian Process Factorization Machines for Context-aware Recommendations Trung V. Nguyen ANU & NICTA Canberra, Australia u5075561@anu.edu.au Alexandros Karatzoglou Telefonica Research Barcelona, Spain
More informationBlack-box α-divergence Minimization
Black-box α-divergence Minimization José Miguel Hernández-Lobato, Yingzhen Li, Daniel Hernández-Lobato, Thang Bui, Richard Turner, Harvard University, University of Cambridge, Universidad Autónoma de Madrid.
More informationNeural Networks and Machine Learning research at the Laboratory of Computer and Information Science, Helsinki University of Technology
Neural Networks and Machine Learning research at the Laboratory of Computer and Information Science, Helsinki University of Technology Erkki Oja Department of Computer Science Aalto University, Finland
More informationECE521 Lecture 19 HMM cont. Inference in HMM
ECE521 Lecture 19 HMM cont. Inference in HMM Outline Hidden Markov models Model definitions and notations Inference in HMMs Learning in HMMs 2 Formally, a hidden Markov model defines a generative process
More informationGaussian Processes. Le Song. Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012
Gaussian Processes Le Song Machine Learning II: Advanced Topics CSE 8803ML, Spring 01 Pictorial view of embedding distribution Transform the entire distribution to expected features Feature space Feature
More informationLecture: Gaussian Process Regression. STAT 6474 Instructor: Hongxiao Zhu
Lecture: Gaussian Process Regression STAT 6474 Instructor: Hongxiao Zhu Motivation Reference: Marc Deisenroth s tutorial on Robot Learning. 2 Fast Learning for Autonomous Robots with Gaussian Processes
More informationRanking-Oriented Evaluation Metrics
Ranking-Oriented Evaluation Metrics Weike Pan College of Computer Science and Software Engineering Shenzhen University W.K. Pan (CSSE, SZU) Ranking-Oriented Evaluation Metrics 1 / 21 Outline 1 Introduction
More informationLinear Dynamical Systems
Linear Dynamical Systems Sargur N. srihari@cedar.buffalo.edu Machine Learning Course: http://www.cedar.buffalo.edu/~srihari/cse574/index.html Two Models Described by Same Graph Latent variables Observations
More informationProbability and Information Theory. Sargur N. Srihari
Probability and Information Theory Sargur N. srihari@cedar.buffalo.edu 1 Topics in Probability and Information Theory Overview 1. Why Probability? 2. Random Variables 3. Probability Distributions 4. Marginal
More informationProbabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms
Probabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms François Caron Department of Statistics, Oxford STATLEARN 2014, Paris April 7, 2014 Joint work with Adrien Todeschini,
More informationTemporal Collaborative Filtering with Bayesian Probabilistic Tensor Factorization
Temporal Collaborative Filtering with Bayesian Probabilistic Tensor Factorization Liang Xiong Xi Chen Tzu-Kuo Huang Jeff Schneider Jaime G. Carbonell Abstract Real-world relational data are seldom stationary,
More informationFrom Mating Pool Distributions to Model Overfitting
From Mating Pool Distributions to Model Overfitting Claudio F. Lima 1 Fernando G. Lobo 1 Martin Pelikan 2 1 Department of Electronics and Computer Science Engineering University of Algarve, Portugal 2
More informationDownloaded 09/30/17 to Redistribution subject to SIAM license or copyright; see
Latent Factor Transition for Dynamic Collaborative Filtering Downloaded 9/3/17 to 4.94.6.67. Redistribution subject to SIAM license or copyright; see http://www.siam.org/journals/ojsa.php Chenyi Zhang
More informationSTA414/2104 Statistical Methods for Machine Learning II
STA414/2104 Statistical Methods for Machine Learning II Murat A. Erdogdu & David Duvenaud Department of Computer Science Department of Statistical Sciences Lecture 3 Slide credits: Russ Salakhutdinov Announcements
More informationCollaborative Filtering via Different Preference Structures
Collaborative Filtering via Different Preference Structures Shaowu Liu 1, Na Pang 2 Guandong Xu 1, and Huan Liu 3 1 University of Technology Sydney, Australia 2 School of Cyber Security, University of
More informationSTA 4273H: Sta-s-cal Machine Learning
STA 4273H: Sta-s-cal Machine Learning Russ Salakhutdinov Department of Computer Science! Department of Statistical Sciences! rsalakhu@cs.toronto.edu! h0p://www.cs.utoronto.ca/~rsalakhu/ Lecture 2 In our
More informationNetBox: A Probabilistic Method for Analyzing Market Basket Data
NetBox: A Probabilistic Method for Analyzing Market Basket Data José Miguel Hernández-Lobato joint work with Zoubin Gharhamani Department of Engineering, Cambridge University October 22, 2012 J. M. Hernández-Lobato
More informationA Novel Click Model and Its Applications to Online Advertising
A Novel Click Model and Its Applications to Online Advertising Zeyuan Zhu Weizhu Chen Tom Minka Chenguang Zhu Zheng Chen February 5, 2010 1 Introduction Click Model - To model the user behavior Application
More informationCS249: ADVANCED DATA MINING
CS249: ADVANCED DATA MINING Recommender Systems Instructor: Yizhou Sun yzsun@cs.ucla.edu May 17, 2017 Methods Learnt: Last Lecture Classification Clustering Vector Data Text Data Recommender System Decision
More informationDensity Propagation for Continuous Temporal Chains Generative and Discriminative Models
$ Technical Report, University of Toronto, CSRG-501, October 2004 Density Propagation for Continuous Temporal Chains Generative and Discriminative Models Cristian Sminchisescu and Allan Jepson Department
More informationBayesian Semi-supervised Learning with Deep Generative Models
Bayesian Semi-supervised Learning with Deep Generative Models Jonathan Gordon Department of Engineering Cambridge University jg801@cam.ac.uk José Miguel Hernández-Lobato Department of Engineering Cambridge
More informationMachine Learning for Data Science (CS4786) Lecture 24
Machine Learning for Data Science (CS4786) Lecture 24 Graphical Models: Approximate Inference Course Webpage : http://www.cs.cornell.edu/courses/cs4786/2016sp/ BELIEF PROPAGATION OR MESSAGE PASSING Each
More informationIntroduction to Logistic Regression
Introduction to Logistic Regression Guy Lebanon Binary Classification Binary classification is the most basic task in machine learning, and yet the most frequent. Binary classifiers often serve as the
More informationPattern Recognition and Machine Learning
Christopher M. Bishop Pattern Recognition and Machine Learning ÖSpri inger Contents Preface Mathematical notation Contents vii xi xiii 1 Introduction 1 1.1 Example: Polynomial Curve Fitting 4 1.2 Probability
More informationBayesian Low-Rank Determinantal Point Processes
Bayesian Low-Rank Determinantal Point Processes Mike Gartrell Microsoft mike.gartrell@acm.org Ulrich Paquet Microsoft ulripa@microsoft.com Noam Koenigstein Microsoft noamko@microsoft.com ABSTRT Determinantal
More informationMaximum Margin Matrix Factorization for Collaborative Ranking
Maximum Margin Matrix Factorization for Collaborative Ranking Joint work with Quoc Le, Alexandros Karatzoglou and Markus Weimer Alexander J. Smola sml.nicta.com.au Statistical Machine Learning Program
More informationNatural Gradients via the Variational Predictive Distribution
Natural Gradients via the Variational Predictive Distribution Da Tang Columbia University datang@cs.columbia.edu Rajesh Ranganath New York University rajeshr@cims.nyu.edu Abstract Variational inference
More informationOffline Evaluation of Ranking Policies with Click Models
Offline Evaluation of Ranking Policies with Click Models Shuai Li The Chinese University of Hong Kong Joint work with Yasin Abbasi-Yadkori (Adobe Research) Branislav Kveton (Google Research, was in Adobe
More informationOptimizing Factorization Machines for Top-N Context-Aware Recommendations
Optimizing Factorization Machines for Top-N Context-Aware Recommendations Fajie Yuan 1(B), Guibing Guo 2, Joemon M. Jose 1,LongChen 1, Haitao Yu 3, and Weinan Zhang 4 1 University of Glasgow, Glasgow,
More informationLarge-scale Collaborative Ranking in Near-Linear Time
Large-scale Collaborative Ranking in Near-Linear Time Liwei Wu Depts of Statistics and Computer Science UC Davis KDD 17, Halifax, Canada August 13-17, 2017 Joint work with Cho-Jui Hsieh and James Sharpnack
More informationOnline Advertising is Big Business
Online Advertising Online Advertising is Big Business Multiple billion dollar industry $43B in 2013 in USA, 17% increase over 2012 [PWC, Internet Advertising Bureau, April 2013] Higher revenue in USA
More informationSTA 4273H: Statistical Machine Learning
STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Statistics! rsalakhu@utstat.toronto.edu! http://www.utstat.utoronto.ca/~rsalakhu/ Sidney Smith Hall, Room 6002 Lecture 3 Linear
More informationPersonalized Social Recommendations Accurate or Private
Personalized Social Recommendations Accurate or Private Presented by: Lurye Jenny Paper by: Ashwin Machanavajjhala, Aleksandra Korolova, Atish Das Sarma Outline Introduction Motivation The model General
More informationRanking and Filtering
2018 CS420, Machine Learning, Lecture 7 Ranking and Filtering Weinan Zhang Shanghai Jiao Tong University http://wnzhang.net http://wnzhang.net/teaching/cs420/index.html Content of This Course Another ML
More informationCS Lecture 18. Topic Models and LDA
CS 6347 Lecture 18 Topic Models and LDA (some slides by David Blei) Generative vs. Discriminative Models Recall that, in Bayesian networks, there could be many different, but equivalent models of the same
More informationMatrix Factorization and Collaborative Filtering
10-601 Introduction to Machine Learning Machine Learning Department School of Computer Science Carnegie Mellon University Matrix Factorization and Collaborative Filtering MF Readings: (Koren et al., 2009)
More informationProbabilistic Collaborative Representation Learning for Personalized Item Recommendation
Probabilistic Collaborative Representation Learning for Personalized Item Recommendation Aghiles Salah and Hady W. Lauw School of Information Systems Singapore Management University, Singapore {asalah,
More informationIntroduction to Machine Learning Fall 2017 Note 5. 1 Overview. 2 Metric
CS 189 Introduction to Machine Learning Fall 2017 Note 5 1 Overview Recall from our previous note that for a fixed input x, our measurement Y is a noisy measurement of the true underlying response f x):
More informationA Bayesian Treatment of Social Links in Recommender Systems ; CU-CS
University of Colorado, Boulder CU Scholar Computer Science Technical Reports Computer Science Spring 5--0 A Bayesian Treatment of Social Links in Recommender Systems ; CU-CS-09- Mike Gartrell University
More informationarxiv: v2 [cs.ir] 4 Jun 2018
Metric Factorization: Recommendation beyond Matrix Factorization arxiv:1802.04606v2 [cs.ir] 4 Jun 2018 Shuai Zhang, Lina Yao, Yi Tay, Xiwei Xu, Xiang Zhang and Liming Zhu School of Computer Science and
More informationCS 188: Artificial Intelligence Fall 2011
CS 188: Artificial Intelligence Fall 2011 Lecture 20: HMMs / Speech / ML 11/8/2011 Dan Klein UC Berkeley Today HMMs Demo bonanza! Most likely explanation queries Speech recognition A massive HMM! Details
More informationVariational Scoring of Graphical Model Structures
Variational Scoring of Graphical Model Structures Matthew J. Beal Work with Zoubin Ghahramani & Carl Rasmussen, Toronto. 15th September 2003 Overview Bayesian model selection Approximations using Variational
More informationA Modified PMF Model Incorporating Implicit Item Associations
A Modified PMF Model Incorporating Implicit Item Associations Qiang Liu Institute of Artificial Intelligence College of Computer Science Zhejiang University Hangzhou 31007, China Email: 01dtd@gmail.com
More informationCoupled Hidden Markov Models: Computational Challenges
.. Coupled Hidden Markov Models: Computational Challenges Louis J. M. Aslett and Chris C. Holmes i-like Research Group University of Oxford Warwick Algorithms Seminar 7 th March 2014 ... Hidden Markov
More informationData Mining Techniques
Data Mining Techniques CS 622 - Section 2 - Spring 27 Pre-final Review Jan-Willem van de Meent Feedback Feedback https://goo.gl/er7eo8 (also posted on Piazza) Also, please fill out your TRACE evaluations!
More informationBayesian leave-one-out cross-validation for large data sets
1st Symposium on Advances in Approximate Bayesian Inference, 2018 1 5 Bayesian leave-one-out cross-validation for large data sets Måns Magnusson Michael Riis Andersen Aki Vehtari Aalto University mans.magnusson@aalto.fi
More information