Saturday, December 13, 2014

NIPS2014 Poster papers

While I listed the proceedings of NIPS recently, it did not seem to include posters at workshops that are taking place today (and took place yesterday). Here are the ones I could find from the workshop pages, enjoy !

Deep Learning and Representation Learning Workshop: NIPS 2014

cuDNN: Efficient Primitives for Deep Learning (#49) Sharan Chetlur, Cliff Woolley, Philippe Vandermersch, Jonathan Cohen, John Tran, Bryan Catanzaro, Evan Shelhamer
Distilling the Knowledge in a Neural Network (#65) Geoffrey Hinton, Oriol Vinyals, Jeff Dean
Supervised Learning in Dynamic Bayesian Networks (#54) Shamim Nemati, Ryan Adams
Deeply-Supervised Nets (#2) Chen-Yu Lee, Saining Xie, Patrick Gallagher, Zhengyou Zhang, Zhuowen Tu
Unsupervised Feature Learning from Temporal Data (#3) Ross Goroshin, Joan Bruna, Arthur Szlam, Jonathan Tompson, David Eigen, Yann LeCun
Autoencoder Trees (#5) Ozan Irsoy, Ethem Alpaydin
Scheduled denoising autoencoders (#6) Krzysztof Geras, Charles Sutton
Learning to Deblur (#8) Christian Schuler, Michael Hirsch, Stefan Harmeling, Bernhard Schölkopf
A Winner-Take-All Method for Training Sparse Convolutional Autoencoders (#10) Alireza Makhzani, Brendan Frey
"Mental Rotation" by Optimizing Transforming Distance (#11) Weiguang Ding, Graham Taylor
On Importance of Base Model Covariance for Annealing Gaussian RBMs (#12) Taichi Kiwaki, Kazuyuki Aihara
Ultrasound Standard Plane Localization via Spatio-Temporal Feature Learning with Knowledge Transfer (#14) Hao Chen, Dong Ni, Ling Wu, Sheng Li, Pheng Heng
Understanding Locally Competitive Networks (#15) Rupesh Srivastava, Jonathan Masci, Faustino Gomez, Jurgen Schmidhuber
Unsupervised pre-training speeds up the search for good features: an analysis of a simplified model of neural network learning (#18) Avraham Ruderman
Analyzing Feature Extraction by Contrastive Divergence Learning in RBMs (#19) Ryo Karakida, Masato Okada, Shun-ichi Amari
Deep Tempering (#20) Guillaume Desjardins, Heng Luo, Aaron Courville, Yoshua Bengio
Learning Word Representations with Hierarchical Sparse Coding (#21) Dani Yogatama, Manaal Faruqui, Chris Dyer, Noah Smith
Deep Learning as an Opportunity in Virtual Screening (#23) Thomas Unterthiner, Andreas Mayr, Günter Klambauer, Marvin Steijaert, Jörg Wenger, Hugo Ceulemans, Sepp Hochreiter
Revisit Long Short-Term Memory: An Optimization Perspective (#24) Qi Lyu, J Zhu
Locally Scale-Invariant Convolutional Neural Networks (#26) Angjoo Kanazawa, David Jacobs, Abhishek Sharma
Deep Exponential Families (#28) Rajesh Ranganath, Linpeng Tang, Laurent Charlin, David Blei
Techniques for Learning Binary Stochastic Feedforward Neural Networks (#29) Tapani Raiko, mathias Berglund, Guillaume Alain, Laurent Dinh
Inside-Outside Semantics: A Framework for Neural Models of Semantic Composition (#30) Phong Le, Willem Zuidema
Deep Multi-Instance Transfer Learning (#32) Dimitrios Kotzias, Misha Denil, Phil Blunsom, Nando De Freitas
Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models (#33) Ryan Kiros, Ruslan Salakhutdinov, Richard Zemel
Retrofitting Word Vectors to Semantic Lexicons (#34) Manaal Faruqui, Jesse Dodge, Sujay Jauhar, Chris Dyer, Eduard Hovy, Noah Smith
Deep Sequential Neural Network (#35) Ludovic Denoyer, Patrick Gallinari
Efficient Training Strategies for Deep Neural Network Language Models (#71)
Holger Schwenk
Deep Learning for Answer Sentence Selection (#36) Lei Yu, Karl Moritz Hermann, Phil Blunsom, Stephen Pulman
Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition (#37) Max Jaderberg, Karen Simonyan, Andrea Vedaldi, Andrew Zisserman
Learning Torque-Driven Manipulation Primitives with a Multilayer Neural Network (#39) Sergey Levine, Pieter Abbeel
SimNets: A Generalization of Convolutional Networks (#41) Nadav Cohen, Amnon Shashua
Phonetics embedding learning with side information (#44) Gabriel Synnaeve, Thomas Schatz, Emmanuel Dupoux
End-to-end Continuous Speech Recognition using Attention-based Recurrent NN: First Results (#45) Jan Chorowski, Dzmitry Bahdanau, KyungHyun Cho, Yoshua Bengio
BILBOWA: Fast Bilingual Distributed Representations without Word Alignments (#46) Stephan Gouws, Yoshua Bengio, Greg Corrado
Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling (#47) Junyoung Chung, Caglar Gulcehre, KyungHyun Cho, Yoshua Bengio
Reweighted Wake-Sleep (#48) Jorg Bornschein, Yoshua Bengio
Explain Images with Multimodal Recurrent Neural Networks (#51) Junhua Mao, Wei Xu, Yi Yang, Jiang Wang, Alan Yuille
Rectified Factor Networks and Dropout (#53) Djork-Arné Clevert, Thomas Unterthiner, Sepp Hochreiter
Towards Deep Neural Network Architectures Robust to Adversarials (#55) Shixiang Gu, Luca Rigazio
Making Dropout Invariant to Transformations of Activation Functions and Inputs (#56) Jimmy Ba, Hui Yuan Xiong, Brendan Frey
Aspect Specific Sentiment Analysis using Hierarchical Deep Learning (#58) Himabindu Lakkaraju, Richard Socher, Chris Manning
Deep Directed Generative Autoencoders (#59) Sherjil Ozair, Yoshua Bengio
Conditional Generative Adversarial Nets (#60) Mehdi Mirza, Simon Osindero
Analyzing the Dynamics of Gated Auto-encoders (#61) Daniel Im, Graham Taylor
Representation as a Service (#63) Ouais Alsharif, Joelle Pineau, philip bachman
Provable Methods for Training Neural Networks with Sparse Connectivity (#66) Hanie Sedghi, Anima Anandkumar
Trust Region Policy Optimization (#67) John D. Schulman, Philipp C. Moritz, Sergey Levine, Michael I. Jordan, Pieter Abbeel
Document Embedding with Paragraph Vectors (#68) Andrew Dai, Christopher Olah, Quoc Le, Greg Corrado
Backprop-Free Auto-Encoders (#69) Dong-Hyun Lee, Yoshua Bengio
Rate-Distortion Auto-Encoders (#73) Luis Sanchez Giraldo, Jose Principe

OPT2014 Optimization for Machine Learning

A Stochastic PCA Algorithm with an Exponential Convergence Rate - Ohad Shamir
Non-Uniform Stochastic Average Gradient Method for Training Conditional Random Fields - Mark Schmidt, Ann Clifton and Anoop Sarkar
Robust minimum volume ellipsoids and higher-order polynomial level sets - Amir Ali Ahmadi, Dmitry Malioutov and Ronny Luss

Convergence Analysis of ADMM for a Family of Nonconvex Problems - Mingyi Hong, Zhi-Quan Luo and Meisam Razaviyayn
Provable Learning of Overcomplete Latent Variable Models: Semi-supervised and Unsupervised Settings - Majid Janzamin, Anima Anandkumar and Rong Ge
Adaptive Communication Bounds for Distributed Online Learning - Michael Kamp, Mario Boley and Michael Mock
Efficient Training of Structured SVMs via Soft Constraints - Ofer Meshi, Nathan Srebro and Tamir Hazan
Approximate Low-Rank Tensor Learning - Yaoliang Yu, Hao Cheng and Xinhua Zhang
Complexity Issues and Randomization Strategies in Frank-Wolfe Algorithms for Machine Learning - Emanuele Frandi, Ricardo Ñanculef and Johan Suykens
On Iterative Hard Thresholding Methods for High-dimensional M-Estimation - Prateek Jain, Ambuj Tewari and Purushottam Kar
Accelerated Parallel Optimization Methods for Large Scale Machine Learning - Haipeng Luo, Patrick Haffner and Jean-Francois Paiement
A Multilevel Framework for Sparse Inverse Covariance Estimation - Eran Treister, Javier Turek and Irad Yavneh
Fast large-scale optimization by unifying stochastic gradient and quasi-Newton methods - Jascha Sohl-Dickstein, Ben Poole and Surya Ganguli
Distributed Latent Dirichlet Allocation via Tensor Factorization - Furong Huang, Sergiy Matusevych, Animashree Anandkumar, Nikos Karampatziakis and Paul Mineiro
Coresets for the DP-Means Clustering Problem - Olivier Bachem, Mario Lucic and Andreas Krause
RadaGrad: Random Projections for Adaptive Stochastic Optimization - Gabriel Krummenacher and Brian Mcwilliams
Scaling up Lloyd’s algorithm: stochastic and parallel block-wise optimization perspectives - Cheng Tang and Claire Monteleoni
CqBoost : A Column Generation Method for Minimizing the C-Bound - François Laviolette, Mario Marchand and Jean-Francis Roy
Tighter Low-rank Approximation via Sampling the Leveraged Element - Srinadh Bhojanapalli, Prateek Jain and Sujay Sanghavi
S2CD: Semi-Stochastic Coordinate Descent - Jakub Konečný, Zheng Qu and Peter Richtárik
Stochastic Relaxation over the Exponential Family: Second-Order Geometry - Luigi Malagò and Giovanni Pistone
Learning with stochastic proximal gradient - Lorenzo Rosasco, Silvia Villa and Bang Cong Vu
Asynchronous Parallel Block-Coordinate Frank-Wolfe - Yu-Xiang Wang, Veeranjaneyulu Sadhanala, Wei Dai, Willie Neiswanger, Suvrit Sra and Eric Xing
Fast Nonnegative Matrix Factorization with Rank-one ADMM - Dongjin Song, David Meyer and Martin Renqiang Min
Neurally Plausible Algorithms Find Globally Optimal Sparse Codes - Sanjeev Arora, Rong Ge, Tengyu Ma and Ankur Moitra
mS2GD: Mini-Batch Semi-Stochastic Gradient Descent in the Proximal Setting - Jakub Konecny, Jie Liu, Peter Richtárik and Martin Takáč
Randomized Subspace Descent - Rafael Frongillo and Mark Reid
Coordinate descent converges faster with the Gauss-Southwell rule than random selection - Mark Schmidt and Michael Friedlander

Modern Nonparametrics 3: Automating the Learning Pipeline

The Randomized Causation Coefficient. (talk)
David Lopez-Paz, Krikamol Muandet, Benjamin Recht.
[paper]
Influence Functions for Nonparametric Estimation. (spotlight, poster)
Kirthevasan Kandasamy, Akshay Krishnamurthy, Barnabás Póczos, Larry Wasserman, James M. Robins.
[paper]
Are You Still Tuning Hyperparameters? Parameter-free Model Selection and Learning. (talk)
Francesco Orabona.
[paper]
Uncertainty Propagation in Gaussian Process Pipelines. (spotlight, poster)
Andreas C. Damianou, Neil D. Lawrence.
[paper, poster]
Kernel non-parametric tests of relative dependency. (spotlight, poster)
Wacha Bounliphone, Arthur Gretton, Matthew Blaschko.
[paper, spotlight, poster]
Generalized Product of Experts for Automatic and Principled Fusion of Gaussian Process Predictions. (spotlight, poster)
Yanshuai Cao, David J. Fleet.
[paper]
Computable Learning via Metric Measure Theory. (spotlight, poster)
Arijit Das, Achim Tresch.
[paper]
Nonparametric Maximum Margin Similarity for Semi-Supervised Learning. (spotlight, poster)
Yingzhen Yang, Xinqi Chu, Zhangyang Wang, Thomas S. Huang.
[paper]
Learning with Deep Trees. (spotlight, poster)
Giulia DeSalvo, Mehryar Mohri, Umar Syed.
[paper, spotlight, poster]
Theoretical Foundations for Learning Kernels in Supervised Kernel PCA. (spotlight, poster)
Mehryar Mohri, Afshin Rostamizadeh, Dmitry Storcheus.
[paper]
Kernel Selection in Support Vector Machines Using Gram-Matrix Properties. (spotlight, poster)
Roberto Valerio, Ricardo Vilalta.
[paper, spotlight (pdf, pptx), poster (pdf, pptx)]

Optimal transport and machine learning

8:30 - 9:10	Crash-Course in Optimal Transport (Organizers)
9:10 - 10:00	Piotr Indyk Geometric representations of the Earth-Mover Distance

	Coffee Break

10:30 - 11:20	Alexander Barvinok Non-negative matrices with prescribed row and column sums (slides)
11:20 - 11:40	Zheng, Pestilli, Rokem Quantifying error in estimates of human brain fiber directions using the EMD (paper)
11:40 - 12:00	Courty, Flamary, Rakotomamonjy, Tuia Optimal transport for Domain adaptation

	Lunch Break

14:40 - 15:00	Flamary, Courty, Rakotomamonjy, Tuia Optimal transport with Laplacian regularization
15:00 - 15:20	Rabin, Papadakis Non-convex relaxation of optimal transport for color transfer (paper)
15:20 - 15:40	Lescornel, Loubès Estimation of deformations between distributions with the Wasserstein distance (paper)
15:40 - 16:30	Adam Oberman Numerical methods for the optimal transportation problem

	Coffee Break

17:00 - 17:50	Robert McCann Optimal transport: old and new
17:50 - 18:30	Panel Discussion

NIPS Workshop on Discrete and Combinatorial Problems in Machine Learning (DISCML) 2014

Invited Talks

Volkan Cevher: A totally unimodular view of structured sparsity

We describe a simple framework for structured sparse recovery based on convex optimization. We show that many interesting structured sparsity models can be naturally represented by linear matrix inequalities on the support of the unknown parameters, where the constraint matrix has a totally unimodular (TU) structure. For such structured models, tight convex relaxations can be obtained in polynomial time via linear programming. Our modeling framework unifies the prevalent structured sparsity norms in the literature, introduces new interesting ones, and renders their tightness and tractability arguments transparent.
Based on
http://infoscience.epfl.ch/record/202767?ln=en http://infoscience.epfl.ch/record/184981?ln=en
Web:
http://lions.epfl.ch/publications

Tom McCormick: A survey of minimizing submodular functions on lattices

We usually talk about submodular functions on the subsets of a finite set, the so-called Boolean lattice, and there are many applications of them. There has been a lot of progress made in the last thirty years in understanding the complexity of, and algorithms for, submodular function minimization (SFM) on the Boolean lattice.
But there are other applications where more general notions of submodularity are important. Suppose that we take the usual definition of submodularity and replace the notion of intersection and union of subsets by the "meet" and "join" in a finite lattice. For example, we could take a rectangular "box" of integer points in n-space, with meet being component-wise min, and join being component-wise max. Then we would like to generalize the results about SFM from the Boolean lattice to more general lattices.
There has been a lot of work on such questions in the last ten years, and this talk will survey some of this work. We start with the case where the lattice is distributive. Here Birkhoff's Theorem allows a clever reduction from SFM on a distributive lattice to ordinary SFM on the Boolean lattice. We then consider lattices that are "oriented" or "signed" generalizations of the Boolean lattice, leading to "bisubmodularity", where we do have good characterizations and algorithms. We also briefly cover recent efforts to further generalize bisubmodularity to "k-submodularity", and to product lattices of "diamonds".
Finally we come back to lattices of integer points in n-space. We demonstrate that submodularity by itself is not enough to allow for efficient minimization, even if we also assume coordinate-wise convexity. This leads to considering concepts from "discrete convexity" such as "L-natural convexity", where we do gain enough structure to allow for efficient minimization.

Yaron Singer: Adaptive seeding: a new framework for stochastic submodular optimization

In this talk we will introduce a new framework for stochastic optimization called adaptive seeding. The framework was originally designed to enable substantial improvements to influence maximization by leveraging a remarkable structural phenomenon in social networks known as the "friendship paradox" (or "your friends have more friends than you"). At a high level, adaptive seeding is the task of making choices in the present that lead to favorable outcomes in the future, and may be of independent interest to those curious about stochastic optimization, submodular maximization, and machine learning. In the talk we will give a taste to some of the problems that arise in this rich domain. We will discuss key algorithmic ideas, as well as empirical and experimental results.

Sebastian Nowozin: Trade-offs in Structured Prediction

Structured learning and prediction problems often turn out to be challenging computationally. Yet, there are different sources of structure and it pays to examine them more closely. One source is a fixed or given or physical constraint on feasible decisions, for example when an actual physical system is being controlled. Another source is the assumption that by modelling the structured domain through latent variables and constraints provides statistical advantages because a small number of parameters can now describe the target domain more accurately and we can learn more effectively from fewer samples by exposing this structure. A third source of combinatorial structure is often the loss function that is used; a structured loss invariably leads to a structured decision problem. When designing our models we are often free to select trade-offs in terms of model structure, capacity, number of parameters, and difficulty of inference; in this talk I will argue that by thinking about the above sources of combinatorial structure we can make more sensible trade-offs and I will illustrate these using example applications from computer vision.

Dhruv Batra: Submodular meets Structured: Finding Diverse Subsets in Exponentially-Large Structured Item Sets

Perception problems are hard and notoriously ambiguous. Robust prediction methods in Computer Vision and Natural Language Processing often search for a diverse set of high-quality candidate solutions or proposals. In structured prediction problems, this becomes a daunting task, as the solution space (image labelings, sentence parses, etc.) is exponentially large.
We study greedy algorithms for finding a diverse subset of solutions in structured-output spaces by drawing new connections between submodular functions over combinatorial item sets and High-Order Potentials (HOPs) studied for graphical models. Specifically, we show that when marginal gains of submodular diversity functions allow structured representations, this enables efficient (sub-linear time) approximate maximization by reducing the greedy augmentation step to inference in a factor graph with appropriately constructed HOPs.
Joint work with Adarsh Prasad (UT-Austin) and Stefanie Jegelka (UC Berkeley).

Representation and Learning Methods for Complex Outputs

08:30--09:15 Invited Talk: Honglak Lee.
Multimodal Deep Learning with Implicit Output Representations
09:15--10:00 Invited Talk: Francesco Dinuzzo.
Output Kernel Learning with Structural Constraints
10:00--10:30 Coffee Break

Session 2
10:30--11:15 Invited Talk: Hal Daume III.
Structured Latent Representations in NLP
11:15--11:30 Contributed Talk: Corinna Cortes, Vitaly Kuznetsov and Mehryar Mohri.
On-line Learning Approach to Ensemble Methods for Structured Prediction
11:30--12:00 Spotlights
Hanchen Xiong, Sandor Szedmak and Justus Piater.
Implicit Learning of Simpler Output Kernels for Multi-Label Prediction
Fabio Massimo Zanzotto and Lorenzo Ferrone.
Output Distributed Convolution Structure Learning for Syntactic Analysis of Natural Language
François Laviolette, Emilie Morvant, Liva Ralaivola and Jean-Francis Roy.
On Generalizing the C-Bound to the Multiclass and Multilabel Settings
Hongyu Guo, Xiaodan Zhu and Martin Renqiang Min.
A Deep Learning Model for Structured Outputs with High-order Interaction
Thomas Unterthiner, Andreas Mayr, Günter Klambauer, Marvin Steijaert, Hugo Ceulemans, Jörg Wenger and Sepp Hochreiter.
Deep Learning for Drug Target Prediction
12:00--14:00 Lunch Break
14:00--15:00 Poster Session

Session 3
15:00--15:45 Invited Talk: Jia Deng.
Learning Visual Models with a Knowledge Graph
15:45--16:00 Contributed Talk: Calvin Murdock and Fernando De La Torre.
Semantic Component Analysis
16:00--16:15 Contributed Talk: Jiaqian Yu and Matthew Blaschko.
Lova ́sz Hinge for Learning Submodular Losses
16:15--16:30 Invited Talk: Rich Sutton.
Representation Learning in Reinforcement Learning
16:30--17:00 Coffee Break

Session 4
17:00--17:45 Invited Talk: Noah Smith.
Loss Functions and Regularization for Less-than-Supervised NLP
17:45--18:00 Contributed Talk: Hichem Sahbi.
Finite State Machines for Structured Scene Decoding
18:00--18:15 Contributed Talk: Luke Vilnis, Nikos Karampatziakis and Paul Mineiro.
Generalized Eigenvectors for Large Multiclass Problems

Distributed Machine Learning and Matrix Computations

Session 1
========
08:15-08:30 Introduction, Reza Zadeh
08:30-09:00 Ameet Talwalkar, MLbase: Simplified Distributed Machine Learning
09:00-09:30 David Woodruff, Principal Component Analysis and Higher Correlations for Distributed Data
09:30-10:00 Virginia Smith, Communication-Efficient Distributed Dual Coordinate Ascent

10:00-10:30 Coffee Break

Session 2
========
10:30-11:30 Jeff Dean (Keynote), Techniques for Training Neural Networks Quickly
11:30-12:00 Reza Zadeh, Distributing the Singular Value Decomposition with Spark
12:00-12:30 Jure Leskovec, In-memory graph analytics

12:30-14:30 Lunch Break

Session 3
========
14:30-15:00 Carlos Guestrin, SFrame and SGraph: Scalable, Out-of-Core, Unified Tabular and Graph Processing
15:00-15:30 Inderjit Dhillon, NOMAD: A Distributed Framework for Latent Variable Models
15:30-16:00 Ankur Dave, GraphX: Graph Processing in a Distributed Dataflow Framework
16:00-16:30 Jeremy Freeman, Large-scale decompositions of brain activity

Minerva: A Scalable and Highly Efficient Training Platform for Deep Learning
Minjie Wang, Tianjun Xiao, Jianpeng Li, Jiaxing Zhang, Chuntao Hong, Zheng Zhang
Maxios: Large Scale Nonnegative Matrix Factorization for Collaborative Filtering
Simon Shaolei Du, Boyi Chen, Yilin Liu, Lei Li
Factorbird - a Parameter Server Approach to Distributed Matrix Factorization
Sebastian Schelter, Venu Satuluri, Reza Zadeh
Improved Algorithms for Distributed Boosting
Jeff Cooper, Lev Reyzin
Parallel and Distributed Inference in Coupled Tensor Factorization Models supplementary
Umut Simsekli, Beyza Ermis, Ali Taylan Cemgil, Figen Oztoprak, S. Ilker Birbil
Dogwild! — Distributed Hogwild for CPU and GPU
Cyprien Noel, Simon Osindero
Generalized Low Rank Models
Madeleine Udell, Corinne Horn, Reza Zadeh, Stephen Boyd
Elastic Distributed Bayesian Collaborative Filtering
Alex Beutel, Markus Weimer, Tom Minka, Yordan Zaykov, Vijay Narayanan
LOCO: Distributing Ridge Regression with Random Projections
Brian McWilliams, Christina Heinze, Nicolai Meinshausen, Gabriel Krummenacher, Hastagiri P. Vanchinathan
Logistic Matrix Factorization for Implicit Feedback Data
Christopher C. Johnson
Tighter Low-rank Approximation via Sampling the Leveraged Element
Srinadh Bhojanapalli, Prateek Jain, Sujay Sanghavi

Networks: From Graphs to Rich Data

NIPS 2014 Workshop

Accepted Papers

Stable Overlapping Replicator Dynamics for Subnetwork Identification Bernard Ng, Burak Yoldemir and Rafeef Abugharbieh
Metric recovery from directed unweighted graphs Tatsunori Hashimoto, Yi Sun and Tommi Jaakkola
Predicting Network Centralities from Node Attributes Harold Soh
The Bayesian Echo Chamber: Modeling Influence in Conversations Fangjian Guo, Charles Blundell, Hanna Wallach and Katherine Heller
Modeling Adoption of Competing Products and Conventions in Social Media Isabel Valera, Manuel Gomez Rodriguez and Krishna Gummadi
Inferring Polyadic Events With Poisson Tensor Factorization Aaron Schein, John Paisley, David M. Blei and Hanna Wallach
Hypernode Graphs for Learning from Binary Relations between Groups in Networks Thomas Ricatte, Remi Gilleron and Marc Tommasi
Scalable Nonparametric Multiway Data Analysis Shandian Zhe, Zenglin Xu, Xinqi Chu, Yuan Qi and Youngja Park
Statistical Models for Degree Distributions of Networks Kayvan Sadeghi and Alessandro Rinaldo
Targeted Influence Maximization through a Social Network Rama Kumar Pasumarthi, Ramasuri Narayanam and Balaraman Ravindran
Refining the Semantics of Social Influence Katerina Marazopoulou, David Arbour and David Jensen
Convex Relaxation of URU^T Tensor Factorization Raphael Bailly, Antoine Bordes and Nicolas Usunier
Random walk models of sparse graphs and networks Benjamin Reddy and Peter Orbanz
Learning to Generate Networks James Atwood, Don Towsley, Krista Gile and David Jensen
Power-Law Graph Cuts Xiangyang Zhou, Jiaxin Zhang and Brian Kulis
On semidefinite relaxations for the block model Arash Amini and Elizaveta Levina
Group-Corrected Stochastic Blockmodels for Community Detection on Large-scale Networks Lijun Peng and Luis Carvalho
Entropy Dynamics of Community Alignment in the Italian Parliament Time-Dependent Network Gabriele Lami, Marco Cristoforetti, Giuseppe Jurman, Cesare Furlanello and Tommaso Furlanello
A unified view of generative models for networks: models, methods, opportunities, and challenges Abigail Jacobs and Aaron Clauset
Unsupervised Induction of Signed Social Networks from Content and Structure Vinodh Krishnan and Jacob Eisenstein

Representation and Learning Methods for Complex Outputs

Flexible Transfer Learning under Support and Model Shift.

Xuezhi Wang and Jeff Schneider

A Probabilistic Model for Dirty Multi-task Feature Selection

Daniel Hernández-Lobato, José Miguel Hernández-Lobato and Zoubin Ghahramani.

Support Consistency of direct Sparse-Change Learning in Markov Networks

Song Liu, Taiji suzuki and Masashi Sugiyama

Deep Learning as an Opportunity in Virtual Screening

Thomas Unterthiner, Andreas Mayr, Günter Klambauer, Marvin Steijaert, Jörg Wenger, Hugo Ceulemans and Sepp Hochreiter

An Improvement to the Domain Adaptation Bound in a PAC-Bayesian context

Pascal Germain, Amaury Habrard, François Laviolette and Emilie Morvant

Mind the Gap: Subspace based Hierarchical Domain Adaptation

Anant Raj, Vinary Namboodiri and Tinne Tuytelaars

Learning from Small Amounts of Labeled data in a Brain Tumor Classification Task

Michael Goetz, Christian Weber, Bram Stieltjes and Klaus Maier-Hein

Active Nearest Neighbors in Changing environments

Christopher Berlind, Ruth Urner

Domain-Adversarial Neural Networks

Hana Ajakan, Pascal Germain, Hugo Larochelle, François Laviolette and Mario Marchand

Data-Effiicient Temporal Regression with Multitask Recurrent Neural Networks.

Sigurd Spieckermann, Steffen Udluft and Thomas Runkler

Sparse Multi-task Reinforcement Learning

Daniele calndirello, Alessandro Lazaric and Marcello Restelli

Multi-task Linear Bandits

Marta Soare, Ouais Alsharif, Alsessandro Lazaric and Joelle Pineau

Bayesian Multitask Distance Metric Learning

Piyush Rai, Wenzhao Lian and Lawrence Carin

Learning Unbiased features

Yujia Li, Kevin Swersky and Richard Zernel

Forecasting Non-Stationary Time Series: From Theory to Algorithms

Vitaly Kuznetsov and Mehryar Mohri

AKBC 2014

4th Workshop on Automated Knowledge Base Construction (AKBC) 2014

Contributed Talks
Arvind Neelakantan, Benjamin Roth and Andrew Mccallum. Knowledge Base Completion using Compositional Vector Space Models PDF
Peter Clark, Niranjan Balasubramanian, Sumithra Bhakthavatsalam, Kevin Humphreys, Jesse Kinkead, Ashish Sabharwal, and Oyvind Tafjord. Automatic Construction of Inference-Supporting Knowledge Bases PDF
Ari Kobren, Thomas Logan, Siddarth Sampangi and Andrew McCallum. Domain Specific Knowledge Base Construction via Crowdsourcing PDF
Posters
Adrian Benton, Jay Deyoung, Adam Teichert, Mark Dredze, Benjamin Van Durme, Stephen Mayhew and Max Thomas. Faster (and Better) Entity Linking with Cascades PDF
Ning Gao, Douglas Oard and Mark Dredze. A Test Collection for Email Entity Linking PDF
Derry Tanti Wijaya, Ndapandula Nakashole and Tom Mitchell. Mining and Organizing a Resource of State-changing Verbs PDF
Jakob Huber, Christian Meilicke and Heiner Stuckenschmidt. Applying Markov Logic for Debugging Probabilistic Temporal Knowledge Bases PDF
Ramanathan Guha. Correlating Entities PDF
Tomasz Tylenda, Sarath Kumar Kondreddi and Gerhard Weikum. Spotting Knowledge Base Facts in Web Texts PDF
Bhavana Dalvi and William Cohen. Multi-view Exploratory Learning for AKBC Problems PDF
Alexander G. Ororbia Ii, Jian Wu and C. Lee Giles. CiteSeerX: Intelligent Information Extraction and Knowledge Creation from Web-Based Data PDF
Adam Grycner, Gerhard Weikum, Jay Pujara, James Foulds and Lise Getoor. A Unified Probabilistic Approach for Semantic Clustering of Relational Phrases PDF
Ndapandula Nakashole and Tom M. Mitchell. Micro Reading with Priors: Towards Second Generation Machine Readers PDF
Edouard Grave. Weakly supervised named entity classification PDF
Lucas Sterckx, Thomas Demeester, Johannes Deleu and Chris Develder. Using Semantic Clustering and Active Learning for Noise Reduction in Distant Supervision PDF
Francis Ferraro, Max Thomas, Matthew Gormley, Travis Wolfe, Craig Harman and Benjamin Van Durme. Concretely Annotated Corpora PDF
Luis Galárraga and Fabian M. Suchanek. Towards a Numerical Rule Mining Language PDF
Bishan Yang and Claire Cardie. Improving on Recall Errors for Coreference Resolution PDF
Chandra Sekhar Bhagavatula, Thanapon Noraset and Doug Downey. TextJoiner: On-demand Information Extraction with Multi-Pattern Queries PDF
Benjamin Roth, Emma Strubell, Katherine Silverstein and Andrew Mccallum. Minimally Supervised Event Argument Extraction using Universal Schema PDF
Mathias Niepert and Sameer Singh. Out of Many, One: Unifying Web-Extracted Knowledge Bases PDF
Laura Dietz, Michael Schuhmacher and Simone Paolo Ponzetto. Queripidia: Query-specific Wikipedia Construction PDF
Jay Pujara and Lise Getoor. Building Dynamic Knowledge Graphs PDF

NIPS2014 - Out of the Box: Robustness in High Dimension

Session 1
========
8:30-9:15 Pradeep Ravikumar, "Dirty Statistical Models"
9:15-10:00 Donald Goldfarb, "Low-rank Matrix and Tensor Recovery: Theory and Algorithms"

Session 2
========
10:30-11:15 Alessandro Chiuso, "Robustness issues in Kernel Tuning: SURE vs. Marginal Likelihood"
11:15-12:00 Po-Ling Loh, "Local optima of nonconvex M-estimators"

Session 3 (contributed talks)
========
15:00-15:20 Vassilis Kalofolias, "Matrix Completion on Graphs"

15:20-15:40 Aleksandr Aravkin, "Learning sparse models using general robust losses"

15:40-16:00 Stephen Becker, "Robust Compressed Least Squares Regression"
16:00-16:20 Ahmet Iscen, "Exploiting high dimensionality for similarity search"

Session 4
========
17:00-17:45 Rina Foygel Barber, "Controlling the False Discovery Rate via Knockoffs" (joint with Emmanuel Candes)

18:45-18:30 Noureddine El Karoui, "On high-dimensional robust regression and inefficiency of maximum likelihood methods"

Modern ML + NLP 2014

D. Bahdanau, K. Cho, and Y. Bengio. Neural Machine Translation by Jointly Learning to Align and Translate.
D. Belanger and S. Kakade. Embedding Word Tokens using a Linear Dynamical System.
C. Cortes, V. Kuznetsov, and M. Mohri. Boosting Ensembles of Structured Prediction Rules.
M. Derezinski and K Rohanimanesh. An Information Theoretic Approach to Quantifying Text Interestingness.
A. Hefny, A. Dubey, S.J. Reddi, and G.J. Gordon. Integrating Transition-based and Graph-based Parsing Using Integer Linear Programming.
I.B. Fidaner and A.T. Cemgil. Clustering Words by Projection Entropy.
G. Forgues, J. Pineau, J.M. Larcheveque, and R. Tremblay. Bootstrapping Dialog Systems with Word Embeddings.
F. Godin, B. Vandersmissen, A. Jalalvand, W. De Neve, and R. Van de Walle. Alleviating Manual Feature Engineering for Part-of-Speech Tagging of Twitter Microposts using Distributed Word Representations.
L.A. Hannah and H.M. Wallach. Summarizing Topics: From Word Lists to Phrases.
C. Hardmeier, J. Tiedemann and J. Nivre. Translating Pronouns with Latent Anaphora Resolution.
X. Liu, K. Duh, T. Iwakura and Y. Matsumoto. Learning Character Representations for Chinese Word Segmentation.
X.V. Lin, S. Singh, L. He, B. Taskar, and L. Zettlemoyer. Multi-Label Learning with Posterior Regularization.
E. Strubell, L. Vilnis, and A. McCallum. Training for Fast Sequential Prediction Using Dynamic Feature Selection.
L. Yeganova, S. Kim, and W.J. Wilbur. Analyzing Medline Topics with a Projection Method.
S. Yoon, S. Lee, and B. Zhang. Predictive Property of Hidden Representations in Recurrent Neural Network Language Models.

Learning Semantics 2014

Machine Reasoning & Artificial Intelligence

08:30a Pedro Domingos, University of Washington, Symmetry-Based Semantic Parsing

08:50a Tomas Mikolov, Facebook, Challenges in Development of Machine Intelligence

09:10a Luke Zettlemoyer, University of Washington, Semantic Parsing for Knowledge Extraction

09:30a Panel Discussion

Contributed Posters

10:00a Contributed Posters, Coffee Break

Natural Language Processing & Semantics from Text Corpora

10:30a Stephen Clark, University of Cambridge, Composition in Distributed Semantics

10:50a Sebastian Riedel, University College London, Embedding Probabilistic Logic for Machine Reading

11:10a Ivan Titov, University of Amsterdam, Inducing Semantics Frames and Roles from Text in a Reconstruction-Error Minimization Framework

11:30a Panel Discussion

Afternoon

Personal Assistants, Dialog Systems, and Question Answering

03:00p Susan Hendrich, Microsoft Cortana

03:20p Ashutosh Saxena, Cornell, Tell Me Dave: Context-Sensitive Grounding of Natural Language into Robotic Tasks

03:40p Jason Weston, Facebook, Memory Networks

04:00p Panel Discussion

Contributed Posters

04:30p Contributed Posters, Coffee Break

Reasoning from Visual Scenes

05:00p Alyosha Efros, UC Berkeley, Towards The Visual Memex

05:20p Jeffrey Siskind, Purdue University, Learning to Ground Sentences in Video

05:40p Larry Zitnick, Microsoft Research, Forget Reality: Learning from Visual Abstraction

06:00p Panel Discussion

Contributed Posters

Morning Session (10:00p-10:30p)

G. Boleda and K. Erk, Distributional semantic features as semantic primitives - or not

C. Burges, E. Renshaw and A. Pastusiak, Relations World: A Possibilistic Graphical Model

J. Cheng, D. Kartsaklis and E. Grefenstette, Investigating the Role of Prior Disambiguation in Deep-learning Compositional Models of Meaning

L. Fagarasan, E. Maria Vecchi and S. Clark, From distributional semantics to feature norms: grounding semantic models in human perceptual data

F. Hill, K. Cho, S. Jean, C. Devin and Y. Bengio, Not all Neural Embeddings are Born Equal

M. Iyyer, J. Boyd-Graber and H. Daumé III, Generating Sentences from Semantic Vector Space Representations

T. Polajnar, L. Rimell and S. Clark, Using Sentence Plausibility to Learn the Semantics of Transitive Verbs

M. Rabinovich and Z. Ghahramani, Efficient Inference for Unsupervised Semantic Parsing

S. Ritter, C. Long, D. Paperno, M. Baroni, M. Botvinick and A. Goldberg, Leveraging Preposition Ambiguity to Assess Representation of Semantic Interaction in CDSM

M. Yu, M. Gormley and M. Dredze, Factor-based Compositional Embedding Models

Afternoon Session (4:10p-5:00p)

J. M. Hernandez Lobato, J. Lloyd, D. Hernandez-Lobato and Z. Ghahramani, Learning the Semantics of Discrete Random Variables: Ordinal or Categorical?

S. J. Hwang and L. Sigal, A Unified Semantic Embedding: Relating Taxonomies and Attributes

Angeliki Lazaridou, Nghia The Pham and Marco Baroni, Combining Language and Vision with a Multimodal Skip-gram Model

M. Malinowski and M. Fritz, Towards a Visual Turing Challenge

G. Synnaeve, M. Versteegh and E. Dupoux, Learning Words from Images and Speech

J. Weston, S. Chopra and A. Bordes, Memory Networks

R. Xu, J. Lu, C. Xiong, J. Corso, Improving Word Representations via Global Visual Context

B. Yang, S. Yih, X. He, J. Gao and L. Deng, Learning Multi-Relational Semantics Using Neural-Embedding Models

Personalization: Methods and Applications