Posted in: AI

IBM Research AI at 2018 Conference on Neural Information Processing Systems

At the Thirty-Second Conference on Neural Information Processing Systems in Montreal, IBM Research AI will share new ideas and results across our portfolio of research aimed at progressing AI towards real-world challenges. Throughout the week, we will present dozens of papers and demos showcasing our work, as listed below. In addition, we will highlight three areas of concentration at our booth 103 that are essential for applying AI more broadly in practice: advancing AI, scaling AI, and trusting AI. Visit to learn about our research, meet our scientists, and try out our latest AI systems and demos including:

Interactive image search based on natural language

IBM scientists developed a novel technique that allows users to search more naturally and effectively for images using natural language. This work improves today’s image search systems where users can search using only a fixed set of attributes. Our novel approach uses relative image captioning and goal-oriented training to create a language-based interface for searching for images such as for items in a retail catalog. Our demo shows how a shopper uses natural language to find the perfect shoe.

Learning to recognize new objects with a few examples

Learning to recognize new objects using a small number of examples is a long-standing challenge in computer vision. IBM Research AI has developed a novel few-shot learning method that learns to accurately recognize new objects from as little as one example. We demonstrate few-shot learning for efficiently training a system to detect logos in natural scenes as well as for recognizing food items for automatic checkout in restaurants.

An open-source library to help detect and remove bias in AI

AI Fairness 360 provides a suite of algorithms for mitigating unwanted bias as well as tutorials for applying them in different domains. The open-source package brings researchers together to aid data scientists, data engineers, and developers in creating and deploying AI solutions that are more fair and accurate. Try it out, or contribute to the toolkit to help engender trust in AI and make the world more equitable for all.

An open platform to accelerate adoption of trusted AI

IBM AI OpenScale allows businesses to automate and operate AI across its full lifecycle, irrespective of how it was built and where it runs. Available later this year, OpenScale will infuse AI with trust and transparency, through toolkits like AI Fairness 360; explain outcomes; mitigate bias; and automate the creation of custom neural networks with NeuNetS (as a beta). Visit our demo to see AI OpenScale in operation.

Automated synthesis of customized neural networks using AI

IBM’s Neural Network Synthesis engine (NeuNetS) automatically synthesizes deep neural networks, which greatly speeds up the development and deployment of accurate deep learning models. As part of IBM AI OpenScale, NeuNetS allows users to build neural networks for specific tasks and datasets in a fraction of the time—without sacrificing accuracy. Create your own neural network in our demo and see how the resulting models can be managed in operation using AI OpenScale.

Quantum computing for AI

Quantum computing is a radically different kind of computing, which has enormous potential to enable discoveries and advances of many kinds— including in AI. While the scale of AI problems we can solve today with quantum computers is modest, we believe we are at the beginning of a journey that will dramatically speed up computations that are at the heart of advancing AI. Our demo shows the implementation of a machine learning classification task using quantum computing.

EXPO and more

In addition to our booth, demos, and papers, at the Conference’s first-ever EXPO on Sunday, December 2, the lead scientists behind IBM Project Debater will give a technical talk about the first AI system to debate complex topics with people (8:00 am in Room 517D). You can also stop by the IBM Research AI booth 103 during the week to speak with them about advancing AI for mastering language. Later in the EXPO, IBM researchers will demonstrate how AI Factsheets improve transparency towards creating more trusted AI services. Come to Room 510ABCD from 2:00 ­– 6:30 pm to see how it works.

IBM Research AI is a Diamond sponsor of the Conference and is also proud to be a Platinum sponsor of both the 13th Women in Machine Learning Workshop (Monday, December 3) and the 2nd Black in AI Workshop (Saturday, December 7).

We look forward to seeing you in Montreal!


Accepted papers

Co-regularized Alignment for Unsupervised Domain Adaptation
Abhishek Kumar, Prasanna Sattigeri, Kahini Wadhawan, Leonid Karlinsky, Rogerio Feris, Bill Freeman, Gregory Wornell
Tue Dec 4, 10:45 am – 12:45 pm, Room 210 & 230 AB #165

Constrained Generation of Semantically Valid Graphs via Regularizing Variational Autoencoders
Tengfei Ma, Jie Chen, Cao Xiao
Tue Dec 4, 10:45 am – 12:45 pm, Room 210 & 230 AB #9

Evolutionary Stochastic Gradient Descent for Optimization of Deep Neural Networks
Xiaodong Cui, Wei Zhang, Zoltan Tuske, Michael Picheny
Tue Dec 4, 10:45 am – 12:45 pm, Room 210 & 230 AB #120

Neural Interaction Transparency (NIT): Disentangling Learned Interactions for Improved Interpretability
Michael Tsang, Hanpeng Liu, Sanjay Purushotham, Pavankumar Murali, Yan Liu
Tue Dec 4, 10:45 am – 12:45 pm, Room 210 & 230 AB #83

Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding
Kexin Yi, Jiajun Wu, Chuang Gan, Antonio Torralba, Pushmeet Kohli, Joshua B. Tenenbaum
Tue Dec 4, 3:40 – 3:45 pm, Room 220 CD (Spotlight)
Tue Dec 4, 5:00 – 7:00 pm, Room 210 & 230 AB #21

Dialog-based Interactive Image Retrieval
Xiaoxiao Guo, Hui Wu, Yu Cheng, Steven Rennie, Gerald Tesauro, Rogerio Feris
Tue Dec 4, 5:00 – 7:00 pm, Room 210 & 230 AB #55

Explanations based on the Missing: Towards Contrastive Explanations with Pertinent Negatives
Amit Dhurandhar, Pin-Yu Chen, Ronny Luss, Chun-Chen Tu, Paishun Ting, Karthikeyan Shanmugam, Payel Das
Tue Dec 4, 5:00 – 7:00 pm, Room 210 & 230 AB #127

MiME: Multilevel Medical Embedding of Electronic Health Records for Predictive Healthcare
Edward Choi, Cao Xiao, Walter F. Stewart, Jimeng Sun
Tue Dec 4, 5:00 – 7:00 pm, Room 210 & 230 AB #102

Snap ML: A Hierarchical Framework for Machine Learning
Celestine Dünner, Thomas Parnell, Dimitrios Sarigiannis, Nikolas Ioannou, Andreea Anghel, Haralampos Pozidis
Tue Dec 4, 5:00 – 7:00 pm, Room 210 & 230 AB #109

Training Deep Neural Networks with 8-bit Floating Point Numbers
Naigang Wang, Jungwook Choi, Daniel Brand, Chia-Yu Chen, Kailash Gopalakrishnan
Tue Dec 4, 5:00 – 7:00 pm, Room 210 & 230 AB #108

Delta-Encoder: An Effective Sample Synthesis Method for Few-Shot Object Recognition
Eli Schwartz, Leonid Karlinsky, Joseph Shtok, Sivan Harary, Mattias Marder, Rogerio Feris, Abhishek Kumar, Raja Giryes, Alex M. Bronstein
Wed Dec 5, 9:50 – 9:55 am, Room 220 E (Spotlight)
Wed Dec 5, 10:45 am – 12:45 pm, Room 210 & 230 AB #25

Efficient Neural Network Robustness Certification with General Activation Functions
Huan Zhang, Tsui-Wei Weng, Pin-Yu Chen, Cho-Jui Hsieh, Luca Daniel
Wed Dec 5, 10:45 am – 12:45 pm, Room 210 & 230 AB #147

Weakly Supervised Dense Event Captioning in Videos
Xuguang Duan, Wenbing Huang, Chuang Gan, Jingdong Wang, Wenwu Zhu, Junzhou Huang
Wed Dec 5, 10:45 am – 12:45 pm, Room 210 & 230 AB #125

From Stochastic Planning to Marginal MAP
Hao Cui, Radu Marinescu, Roni Khardon
Wed Dec 5, 5:00 – 7:00 pm, Room 210 & 230 AB #169

Gradient Descent for Spiking Neural Networks
Dongsung Huh, Terrence J. Sejnowski
Wed Dec 5, 5:00 – 7:00 pm, Room 210 & 230 AB #69

Improving Simple Models with Confidence Profiles
Amit Dhurandhar, Karthikeyan Shanmugam, Ronny Luss, Peder Olsen
Wed Dec 5, 5:00 – 7:00 pm, Room 210 & 230 AB #90

Learning Abstract Options
Matthew Riemer, Miao Liu, Gerald Tesauro
Wed Dec 5, 5:00 – 7:00 pm, Room 210 & 230 AB #145

Proximal Graphical Event Models
Debarun Bhattacharjya, Dharmashankar Subramanian, Tian Gao
Wed Dec 5, 5:00 – 7:00 pm, Room 210 & 230 AB #6

Zeroth-Order Stochastic Variance Reduction for Nonconvex Optimization
Sijia Liu, Bhavya Kailkhura, Pin-Yu Chen, Paishun Ting, Shiyu Chang, Lisa Amini
Wed Dec 5, 5:00 – 7:00 pm, Room 210 & 230 AB #51

Boolean Decision Rules via Column Generation
Sanjeeb Dash, Oktay Gunluk, Dennis Wei
Thu Dec 6, 9:45 – 9:50 am, Room 517 CD (Spotlight)
Thu Dec 6, 10:45 am – 12:45 pm, Room 210 & 230 AB #79

Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distributions
Sara Magliacane, Thijs van Ommen, Tom Claassen, Stephan Bongers, Philip Versteeg , Joris M. Mooij
Thu Dec 6, 10:45 am – 12:45 pm, Room 210 & 230 AB #3

Experimental Design for Cost-Aware Learning of Causal Graphs
Erik M. Lindgren, Murat Kocaoglu, Alexandros G. Dimakis, Sriram Vishwanath
Thu Dec 6, 10:45 am – 12:45 pm, Room 210 & 230 AB #1

On Controllable Sparse Alternatives to Softmax
Anirban Laha, Saneem A. Chemmengath, Priyanka Agrawal, Mitesh M. Khapra, Karthik Sankaranarayanan, Harish G. Ramaswamy
Thu Dec 6, 5:00 – 7:00 pm, Room 210 & 230 AB #65

The Limits of Post-Selection Generalization
Kobbi Nissim, Adam Smith, Thomas Steinke, Uri Stemmer, Jonathan Ullman
Thu Dec 6, 5:00 – 7:00 pm, Room 210 & 230 AB #146

Workshop papers

Rigorous Analysis of Racial Bias in Gender Classification
Vidya Muthukumar, Tejaswini Pedapati, Nalini Ratha, Prasanna Sattigeri, Chai-Wah Wu, Brian Kingsbury, Abhishek Kumar, Samuel Thomas, Aleksandra Mojsilovic, Kush R. Varshney
Women in Machine Learning Workshop
Mon Dec 3, 9:00 am

Unsupervised Learning with Contrastive Latent Variable Models
Kristen Severson, Soumya Ghosh, Kenney Ng
Women in Machine Learning Workshop
Mon Dec 3, 9:00 am

Improving Viseme Recognition with Synthetic Data using CNNs and GAN-based Multi-view Mapping
Andrea Britto Mattos Lima, Dario Augusto Borges Oliveira, Edmilson da Silva Morais
Women in Machine Learning Workshop
Mon Dec 3, 9:00 am

Can We Train Multi-Party Turn Taking Models from Dialogues Logs?
Maíra Gatti de Bayser, Paulo Cavalin, Claudio Pinhanez, Bianca Zadrozny
Women in Machine Learning Workshop
Mon Dec 3, 9:00 am

Explaining Lung Nodules through Semantic and Numerical Features
Maysa Malfiza Garcia de Macedo, Dario Augusto Borges Oliveira
Women in Machine Learning Workshop
Mon Dec 3, 9:00 am

On Controllable Sparse Alternatives to Softmax
Anirban Laha, Saneem Ahmed Chemmengath, Priyanka Agrawal, Mitesh Khapra, Karthik Sankaranarayanan, Harish Ramaswamy
Women in Machine Learning Workshop
Mon Dec 3, 9:00 am

Merging Datasets Through Deep Learning
Kavitha Srinivas, Abraham Gale, Julian Dolby
Women in Machine Learning Workshop
Mon Dec 3, 9:00 am

Deep Nested Hierarchical Dirichlet Processes
Priyanka Agrawal, Lavanya Tekumalla, Indrajit Bhattacharya
Women in Machine Learning Workshop
Mon Dec 3, 9:00 am

Snap ML: A Hierarchical Framework for Machine Learning
Celestine Dünner, Thomas Parnell, Dimitrios Sarigiannis, Nikolas Ioannou, Andreea Anghel, Gummadi Ravi, Madhusudanan Kandasamy, Haralampos Pozidis
Women in Machine Learning Workshop
Mon Dec 3, 9:00 am

Practical Considerations for Probabilistic Backpropagation
Matthew Benatan, Edward O. Pyzer-Knapp
Bayesian Deep Learning
Fri Dec 7, 8:00 am – 6:30 pm

Probabilistic Mixture of Model Agnostic Meta Learners
Prasanna Sattigeri, Soumya Ghosh, Abhishek Kumar, Karthikeyan Ramamurthy, Samuel Hoffman, Youssef Drissi, Inkit Padhi
Bayesian Deep Learning
Fri Dec 7, 8:00 am – 6:30 pm

Deep Nested Hierarchical Dirichlet Processes
Priyanka Agrawal, Lavanya Tekumalla, Indrajit Bhattacharya
Bayesian Deep Learning
Fri Dec 7, 8:00 am – 6:30 pm

Latent Projection BNNs: Avoiding Weight-Space Pathologies by Projecting Neural Network Weights
Weiwei Pan, Melanie Fernandez Pradier, Jiayu Yao, Finale Doshi-Velez, Soumya Ghosh
Bayesian Deep Learning
Fri Dec 7, 8:00 am – 6:30 pm

Knowledge Grounded End-to-End Dialog
Lazaros C. Polymenakos
2nd Conversational AI Workshop: Today’s Practice and Tomorrow’s Potential
Fri Dec 7, 8:00 am – 6:30 pm

A Bandit Approach to Posterior Dialog Orchestration Under a Budget
Sohini Upadhyay, Mayank Agarwal, Djallel Bouneffouf, Yasaman Khazaeni
2nd Conversational AI Workshop: Today’s Practice and Tomorrow’s Potential
Fri Dec 7, 8:00 am – 6:30 pm

An Interpretable Machine Learning Methodology for Well Data Integration and Sweet Spotting Identification
Jorge Luis Guevara Diaz
Machine Learning for Geophysical & Geochemical Signals
Fri Dec 7, 8:00 am – 6:30 pm

Learning Beyond Simulated Physics
Alexis Asseman, Tomasz Kornuta, Ahmet Ozcan
Modeling and Decision-Making in the Spatiotemporal Domain
Fri Dec 7, 8:00 am – 6:30 pm

Deep Nested Hierarchical Dirichlet Processes
Priyanka Agrawal, Lavanya Tekumalla, Indrajit Bhattacharya
All of Bayesian Nonparametrics (Especially the Useful Bits)
Fri Dec 7, 8:00 am – 6:30 pm

Sampling Acquisition Functions for Batch Bayesian Optimization
Alessandro De Palma, Celestine Dünner, Thomas Parnell, Andreea Anghel, Haralampos Pozidis
All of Bayesian Nonparametrics (Especially the Useful Bits)
Fri Dec 7, 8:00 am – 6:30 pm

Elastic CoCoA: Scaling In to Improve Convergence
Michael Kaufmann, Thomas Parnell, Kornilios Kourtis
MLSys: Workshop on Systems for ML and Open Source Software
Fri Dec 7, 8:00 am – 6:30 pm

Parallel Training of Linear Models without Compromising Convergence
Nikolas Ioannou, Celestine Dünner, Kornilios Kourtis, Thomas Parnell
MLSys: Workshop on Systems for ML and Open Source Software
Fri Dec 7, 8:00 am – 6:30 pm

Benchmarking and Optimization of Gradient Boosting Decision Tree Algorithms
Andreea Anghel, Nikolaos Papandreou, Thomas Parnell, Alessandro De Palma, Haralampos Pozidis
MLSys: Workshop on Systems for ML and Open Source Software
Fri Dec 7, 8:00 am – 6:30 pm

On Transfer Learning Using a MAC Model Variant
Vincent Marois, Jayram Thathachar, Vincent Albouy, Tomasz Kornuta, Younes Bouhadjar, Ahmet S. Ozcan
Visually Grounded Interaction and Language (VIGIL)
Fri Dec 7, 8:00 am – 6:30 pm

Scalable Graph Learning for Anti-Money Laundering: A First Look
Mark Weber, Jie Chen, Toyotaro Suzumura, Aldo Pareja, Tengfei Ma, Hiroki Kanezashi, Tim Kaler, CHarles E. Leiserson, Tao B. Schardl
Challenges and Opportunities for AI in Financial Services: The Impact of Fairness, Explainability, Accuracy, and Privacy
Fri Dec 7, 8:00 am – 6:30 pm

Continual Learning by Maximizing Transfer and Minimizing Interference
Matthew Reimer, Ignacio Cases, Robert Ajemian, Miao Liu, Irina Rish, Yuhai Tu, Gerald Tesauro
Continual Learning Workshop
Fri Dec 7, 8:00 am – 6:30 pm

Continual Learning with Self-Organizing Maps
Pouya Bashivan, Martin Schrimpf, Robert Ajemian, Irina Rish, Matthew Reimer, Yuhai Tu
Continual Learning Workshop
Fri Dec 7, 8:00 am – 6:30 pm

Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distributions
Sara Magliacane, Thijs van Ommen, Tom Claassen, Stephan Bongers, Philip Versteeg, Joris Mooij
Causal Learning
Fri Dec 7, 8:00 am – 6:30 pm

Entropic Latent Variable Discovery
Murat Kocaoglu, Sanjay Shakkottai, Alexandros G. Dimakis, Constantine Caramanis, Sriram Vishwanath
Causal Learning
Fri Dec 7, 8:00 am – 6:30 pm

Incorporating Attention in World Models for Improved Dynamic Modeling
Deepika Bablani, Parth Chadha
Modeling the Physical World: Learning, Perception, and Control
Fri Dec 7, 8:00 am – 6:30 pm

Model Poisoning Attacks in Federated Learning
Arjun Nitin Bhagoji, Supriyo Chakraborty, Prateek Mittal, Seraphin Calo
Workshop on Security in Machine Learning
Fri Dec 7, 8:45 am – 5:15 pm

COCO-Africa: A Curation Tool and Dataset of Common Objects in the Context of Africa
Victor Dibia
2nd Black in AI Workshop
Fri Dec 7, 1:00 – 6:00 pm

From Node Embedding to Graph Embedding: Scalable Global Graph Kernel via Random Features
Lingfei Wu, Ian En-Hsu Yen, Liang Zhao, Kun Xu, Michael Witbrock
Relational Representation Learning Workshop
Sat Dec 8, 8:00 am – 6:30 pm

Symbolic Relation Networks for Reinforcement Learning
Dhaval Adjodah, Tim Klinger, Josh Joseph
Relational Representation Learning Workshop
Sat Dec 8, 8:00 am – 6:30 pm

Image-Level Attentional Context Modeling Using Nested-Graph Neural Networks
Guillaume Jaume
Relational Representation Learning Workshop
Sat Dec 8, 8:00 am – 6:30 pm

Unsupervised Learning with Contrastive Latent Variable Models
Kristen Severson, Soumya Ghosh, Kenney Ng
Machine Learning for Health (ML4H): Moving Beyond Supervised Learning in Healthcare
Sat Dec 8, 8:00 am – 6:30 pm

Hierarchical Deep Learning Classification of Unstructured Pathology Reports to Automate ICD-O Morphology Grading
Waheeda Saib, Tapiwa Chiwewe
Machine Learning for Health (ML4H): Moving Beyond Supervised Learning in Healthcare
Sat Dec 8, 8:00 am – 6:30 pm

Vox2Net: From 3D Shapes to Network Sculptures & GAN sculpture in NeurIPS Art Gallery
Mauro Martino, Luca Stornaiuolo, Nima Dehmamy
Second Workshop on Machine Learning for Creativity and Design
Sat Dec 8, 8:00 am – 6:30 pm

Molecular Transformer for Chemical Reaction Prediction and Uncertainty Estimation
Philippe Schwaller, Teodoro Laino, Théophile Gaudin, Peter Bolgar, Costas Bekas, Alpha A. Lee
Machine Learning for Molecules and Materials
Sat Dec 8, 8:00 am – 6:30 pm

PepCVAE: Semi-Supervised Targeted Design of Antimicrobial Peptide Sequences
Payel Das, Kahini Wadhawan, Oscar Chang, Tom Sercu, Cicero Dos Santos, Matthew Riemer, Inkit Padhi, Vijil Chenthamarakshan, Aleksandra Mojsilovic
Machine Learning for Molecules and Materials
Sat Dec 8, 8:00 am – 6:30 pm

Powerful, Transferable Representations for Molecules through Intelligent Task Selection in Deep Multitask Networks
Clyde Fare, Lukas Turcani, Edward O. Pyzer-Knapp
Machine Learning for Molecules and Materials
Sat Dec 8, 8:00 am – 6:30 pm

PaccMann: Predicting Anticancer Compound Sensitivity with Multimodal Attention-Based Neural Networks
Ali Oskooei, Jannis Born, Matteo Manica, Vigneshwari Subramanian, Julio Sáez-Rodríguez, María Rodríguez Martínez
Machine Learning for Molecules and Materials
Sat Dec 8, 8:00 am – 6:30 pm

Inference of the Three-Dimensional Chromatin Structure and its Temporal Behavior
Bianca-Cristina Cristescu, Zalán Borsos, John Lygeros, María Rodríguez Martínez, Maria Anna Rapsomaniki
Machine Learning for Molecules and Materials
Sat Dec 8, 8:00 am – 6:30 pm

Accelerating Machine Learning Research with MI-Prometheus
Tomasz Kornuta, Vincent Marois, Ryan L. McAvoy, Younes Bouhadjar, Alexis Asseman, Vincent Albouy, T.S. Jayram, Ahmet S. Ozcan
MLOSS 2018: Sustainable Communities
Sat Dec 8, 8:00 am – 6:30 pm

Open Fabric for Deep Learning Models
Falk Pollock, Scott Boag, Maria-Irina Nicolae
MLOSS 2018: Sustainable Communities
Sat Dec 8, 8:00 am – 6:30 pm

Support Fuzzy-Set Machines: From Kernels on Fuzzy Sets to Machine Learning Applications
Jorge Luis Guevara Diaz
LatinX AI
Sat Dec 8, 8:00 am – 5:00 pm

Information Theoretic Generative Modeling
Luis Lastras
LatinX AI
Sat Dec 8, 8:00 am – 5:00 pm

Workshops

13th Women in Machine Learning Workshop
Mon Dec 3, 9:00 am – 10:00 pm

Competition Track Day 1: TrackML a LSTM attempt
Fri Dec 7, 8:00 am – 6:30 pm

2nd Conversational AI Workshop: Today’s Practice and Tomorrow’s Potential
Fri Dec 7, 8:00 am – 6:30 pm

2nd Black in AI Workshop
Fri Dec 7, 1:00 – 6:00 pm

Demonstrations

Automatic Generation of Factsheets for Trusted AI in a Runtime Environment
Sun Dec 2, 2:00 – 6:30 pm, Room 510ABCD

A Machine Learning Environment to Determine Novel Malaria Policies
Tue Dec 4, 10:45 am – 7:30 pm, Room 510ABCD D9

Game for Detecting Backdoor Attacks on Deep Neural Networks using Activation Clustering
Tue Dec 4, 10:45 am – 7:30 pm, Room 510

BigBlueBot: A Demonstration of How to Detect Egregious Conversations with Chatbots
Wed Dec 5, 10:45 am – 7:30 pm, Room 510ABCD D1

PatentAI: IP Infringement Detection with Enhanced Paraphrase Identification
Wed Dec 5, 10:45 am – 7:30 pm, Room 510ABCD D5

John R. Smith

IBM Fellow, IBM Research