• wiml
  • October 28, 2016

The 2020 WiML Un-Workshop at ICML will be held virtually on Monday, July 13th, 2020.

All participants are required to abide by the WiML Code of Conduct.

Please use this link to access the un-workshop on ICML.


The program will include the following line-up of invited speakers:

2020 Invited Speakers
Monday, July 13th, 2020
Time (GMT/UTC) Event
06:00 – 06:40 Informal Social
06:40 – 06:50 Introduction and opening remarks
06:50 – 07:00 WiML D&I chairs remarks – Sinead Williamson and Rachel Thomas
07:00 – 07:35 Invited talk – Sara van de Geer
07:35 – 08:35 Breakout sessions #1
08:35 – 08:50 Virtual coffee break #1
08:50 – 09:25 Invited talk – Naila Murray
09:25 – 10:25 Breakout sessions #2
10:25 – 11:10 Sponsor Expo: Presentations by QuantumBlack, Netflix and IBM
11:10 – 15:10 Break + informal social
15:10 – 16:10 Sponsor Expo: Presentations by Google, Apple, DeepMind and Facebook
16:10 – 16:45 Invited talk – Nancy Reid
16:45 – 17:45 Breakout sessions #3
17:45 – 18:00 Virtual coffee break #2
18:00 – 18:35 Invited talk – Doina Precup
18:35 – 19:35 Breakout sessions #4
19:35 – 20:35 Mentoring panel – Doina Precup, Deborah Raji, Anima Anandkumar, Angjoo Kanazawa and Sinead Williamson (moderator).
20:35 – 20:50 WiML president remarks – Sarah Osentoski
20:50 – 21:00 Break
21:00 – 23:00 Joint affinity groups poster session


Our sponsor booths are open during the un-workshop. Please find information on their schedules and events here.
For more details about the breakout sessions (for example, affiliations), please use this link.
Please use this link to find out more about invited speakers.


Breakout session #1, 7:35 AM – 8:35 AM GMT

ID Session title Leaders Facilitators
1.1 Inference of Cross-lingual Language Models Gagana Coolga, Stuti Gupta Akash Smaran
1.2 ML4GlobalHealth: Issues and Opportunities in Resource-Limited Settings Mary-Anne Hartley, Danielle Belgrave Berthine Nyunga
1.3 Deep Learning for Natural Language Processing in Low Resource settings Surangika Ranathunga, Rishemjit Kaur, Annie En-Shiun Lee Marjana Skenduli, Mehreen Alam
1.4 AI and Creativity: Generative Art Aneta Neumann, Frehiwot Girmay Aparna Akula, Tina Raissi


Breakout session #2, 9:25AM – 10:25AM GMT

ID Session title Leaders Facilitators
2.1 Well-specified Scalable Models with Variational Inference Ines Krissaane, Samrudhdhi Rangrej Laya Rafiee
2.2 Future of data: How will data diversity become a requirement for training AI models Adepeju Oshisanya, Allison Gardner, Aylin Cakiroglu, Simone Larsson Celine Lature
2.3 AI and Morality: Teaching Human Morality to Machines Frehiwot Girmay, Aneta Neumann Georgie Kennedy, Paula Hall


Breakout session #3, 4:45PM – 5:45PM GMT

ID Session title Leaders Facilitators
3.1 Healthcare and Machine Learning: real world applications and challenges Olga Liakhovich, Tempest van Schaik, Summer Elasady, Bianca Furtuna Katie Claveau
3.2 Recommender System Research in Industry Ghazal Fazelnia, Zahra Nazari, Mozhgan Saeidi Krystal Maughan, Sneha Srinivasan
3.3 Applied Category Theory Tai-Danae Bradley Melanie Weber
3.4 Mining biological and biomedical data with graph-based
Natalie Stanley, Huda Nassar, Ina Stelzer Jolene Ranek
3.5 Feminist Perspectives for Machine Learning & Computer Vision Fatemehsadat Mireshghallah, Srishti Yadav, Mary Anne Smart Jin (Alice) Qixuan
3.6 Tackling Climate Change with Machine Learning Priya Donti, Sasha Luccioni David Rolnick
3.7 Not Just Another Application: Applications for Social Good Jennifer Hobbs, Saba Dadsetan, Naira Hovakimyan Tania Lorido Botran, Lori Liu
3.8 Optimization Challenges of Generative Adversarial Networks Reyhane Askari Hemmat, Alexia Jolicoeur-Martineau, Laya Rafiee Xing Han
3.9 Challenges and practices in deploying AI in Medical Imaging Weiwei Zong, Manju Liu Zhen Sun
3.10 Entangled Conversations on Disentangled Representations (EnCoDR) Chhavi Yadav, Irina Higgins, Jovana Mitrović Laure Delisle, Niveditha Kalavakonda


Breakout session #4, 6:35PM – 7:35PM GMT

ID Session title Leaders Facilitators
4.1 Un-Bookclub: Race After Technology Anoush Najarian, Ishaani, Aleshia Hayes Sindhuja Parimalarangan, Louvere Walker-Hannon
4.2 Fairness and bias in ML and NLP Swetasudha Panda, Emily Black, Xueru Zhang Shikha Bordia
4.3 Coping with sample inefficiency of deep-reinforcement learning (DRL) for embodied AI Vidhi Jain, Simin Liu Ganesh Iyer
4.4 Performative Prediction: When Predictions Impact the Predicted Celestine Mendler-Dünner, Tijana Zrnic Juan Carlos Perdomo
4.5 Robust Machine Learning with Bad Training Data Sergul Aydore, Haleh Akrami Berna Kabadayi
4.6 Machine Learning for Neuroimaging Elvisha Dhamala, Meenakshi Khosla Carmen Khoo
4.7 A Review of Early Exit Training and Inference techniques Vaidheeswaran Archana, Sherin Mathews, Yashika Sharma Zahra Vaseqi
4.8 Continual Reinforcement Learning Khimya Khetarpal, Rose E. Wang, Feryal Behbahani Arundhati Banerjee
4.9 Uncertainty Estimation in Bayesian Deep Learning Polina Kirichenko, Melanie F. Pradier, Weiwei Pan Ana-Denisa Secuiu
4.10 Towards children-aware machine learning with a focus on NLP challenges and applications Belen Saldias, Safinah Ali Tamara Covacevich, Clare Liu


Sponsor Expo Presentations

Time GMT/UTC Sponsor Speaker Title
10:25 – 10:40 QuantumBlack Maren Eckhoff Using AI for social and global good
10:40 – 10:55 Netflix Maria Dimakopoulou Slate Bandit Learning & Evaluation
10:55 – 11:10 IBM Lisa Amini IBM Research AI
15:10 – 15:25 Google Jennifer Wei Machine Learning for Smell: Learning Generalizable Perceptual Representations of Small Molecules
15:25 – 15:40 Apple Lizi Ottens Machine Learning at Apple
15:40 – 15:55 DeepMind Meire Fortunato DeepMind at WiML Un-workshop
15:55 – 16:10 Facebook Kalesha Bullard Learning to Communicate Nonverbally for Embodied Agent Populations


Invited talk: Sara van de Geer

Total variation regularization

Abstract: Consider the classical problem of learning a signal when observed with noise. One way to do this is to expand the signal in terms of basis functions and then try to learn the coefficients. The collection of basis functions is called a dictionary and the approach is sometimes called “synthesis” because the signal is synthesised from the coefficients. Another learning approach, called “analysis”, is based on an l_1 regularization of a linear operator that describes the signal’s structure. As an example one may think of a signal that lives on a graph, and the linear operator describes the change when going from one node to the next in the graph. The sum of the absolute values of the changes is called the total variation of the signal over the graph. A simple special case is the path graph, and a more complicated one is the two-dimensional grid. We will consider the regularized least squares estimator for such examples and also regularization using total variation of higher order discrete derivatives and Hardy Krause total variation. We will introduce the concept “effective sparsity” which is related to the dimensionality of the unknown signal. The regularized least squares estimator will be shown to mimic an oracle that trades off approximation error and “estimation error”, where the latter depends on the effective sparsity.


Invited talk: Naila Murray

Predicting aesthetic appreciation of images

Abstract: Image aesthetics has become an important criterion for visual content curation on social media sites and media content repositories. Previous work on aesthetic prediction models in the computer vision community has focused on aesthetic score prediction or binary image labeling. However, raw aesthetic annotations are in the form of score histograms and provide richer and more precise information than binary labels or mean scores. In this talk I will present recent work at NAVER LABS Europe on the rarely-studied problem of predicting aesthetic score distributions. The talk will cover the large-scale dataset we collected for this problem, called AVA, and will describe the novel deep architecture and training procedure for our score distribution model. Our model achieves state-of-the-art results on AVA for three tasks: (i) aesthetic quality classification; (ii) aesthetic score regression; and (iii) aesthetic score distribution prediction, all while using one model trained only for the distribution prediction task. I will also discuss our proposed method for modifying an image such that its predicted aesthetics changes, and describe how this modification can be used to gain insight into our model.


Invited talk: Nancy Reid

Distributions for parameters

Abstract: There has been considerable recent controversy over the use of p-values and the phrase “statistically significant”, both in subject matter settings and in the statistical literature. One approach to avoiding the dichotomization associated with hypothesis testing is to provide distributions for parameters. A familiar distribution is the posterior density of Bayesian inference, but there are renewed efforts to provide something like probability statements for parameter while avoiding specification of a prior probability. I will discuss the strengths and limitations of these procedures, with special attention to so-called objective Bayesian approaches.


Invited talk: Doina Precup

Building knowledge for AI agents with reinforcement learning

Abstract: Reinforcement learning allows autonomous agents to learn how to act in a stochastic, unknown environment, with which they can interact. Deep reinforcement learning, in particular, has achieved great success in well-defined application domains, such as Go or chess, in which an agent has to learn how to act and there is a clear success criterion. In this talk, I will focus on the potential role of reinforcement learning as a tool for building knowledge representations in AI agents whose goal is to perform continual learning. I will examine a key concept in reinforcement learning, the value function, and discuss its generalization to support various forms of predictive knowledge. I will also discuss the role of temporally extended actions, and their associated predictive models, in learning procedural knowledge. Finally, I will discuss the challenge of how to evaluate reinforcement learning agents whose goal is not just to control their environment, but also to build knowledge about their world.

WiML Accepted Posters in Joint Affinity Poster Session on Gather.Town

Title Authors
Modeling Coherency in Generated Emails by Leveraging Deep Learners Avisha Das* (University of Houston), Rakesh Verma (University of Houston)
Minimizing Interference and Selection Bias in Network Experiment Design Zahra Fatemi* (University of Illinois at Chicago), Elena Zheleva (University of Illinois at Chicago)
Heating Control in Smart Buildings enabled with Reinforcement Learning Anchal Gupta* (Pennsylvania State University) ,Dr. Youakim Badr (Pennsylvania State University), Dr. Robin Qiu (Pennsylvania State University), Dr. Ashkan Negahban (Pennsylvania State University)
CNN/RNN Approaches for Irregular Pediatric Respiratory Motion Detection Chinasa T. Okolo* (Cornell University) and Bharath Hariharan (Cornell University)
A coupled manifold optimization strategy to jointly model the functional connectomics and behavioral data spaces Niharika Shimona D’Souza* (Johns Hopkins University), Mary Beth Nebel (Johns Hopkins School of Medicine), Nicholas Wymbs (Johns Hopkins School of Medicine), Stewart Mostofsky (Johns Hopkins School of Medicine) , Archana Venkataraman (Johns Hopkins University)
A Meta-Learning Approach for Image Classification Architecture Recommendation Tal Lieber* (Ben Gurion University), Loren Tsahalon (Ben Gurion University), Seffi Cohen (Ben Gurion University), Roman Vainshtein (Ben Gurion University), Bracha Shapira (Ben Gurion University)
A Selection Debiasing Approach for Learning-to-rank Systems Zohreh Ovaisi* (University of Illinois at Chicago), Ragib Ahsan (University of Illinois at Chicago), Yifan Zhang (Sun Yat-sen University), Kathryn Vasilaky (California Polytechnic State University), Elena Zheleva (University of Illinois at Chicago)
Active Domain Randomization for Robust Control Qixuan Jin* (Caltech), Guanya Shi (Caltech), Anqi Liu (Caltech), Haosheng Zou (Caltech), Hao Liu (Caltech), Yisong Yue (Caltech)
Automated Identity Verification with Cross-Domain Image Comparisons Manvi Agarwal* (University of Groningen), Noël Lüneburg (Slimmer AI, Groningen), Rineke Verbrugge (University of Groningen)
CapsuleVAE: Capsule Networks for Enhanced Representation Learning with Variational Autoencoders Zahra Moghimi* (Virginia Tech), Ali Taleb Zadeh Kasgari (Virginia Tech), Walid Saad (Virginia Tech)
Characterization of Potential Drug Treatments for COVID-19 using Social Media data and Machine Learning Ramya Tekumalla* (Georgia State University), Juan M Banda (Georgia State University)
Confounding-Robust Policy Evaluation in Infinite-Horizon Reinforcement Learning Nathan Kallus (Cornell University), Angela Zhou* (Cornell University)
RDFFrames: Knowledge Graph Access for Machine Learning Tools Aisha Mohamed (Qatar Computing Research Institute), Ghadeer Abuoda* (HBKU College of Science and Engineering), Abdurrahman Ghanem (Bluescape), Zoi Kaoudi (Technische Universität Berlin), Ashraf Aboulnaga (Qatar Computing Research Institute)
New Insights from Old Data: Multimodal Classification of Schizophrenia using Neural Architecture Search Gagana B* (AI Without Borders)
Dynamic Algorithm for Social Media Troll Detection Maya Srikanth* (Caltech), Anqi Liu (Caltech), Nicholas Adams-Cohen (Caltech), Michael Alvarez (Caltech), Anima Anandkumar (Caltech)
Dynamic Prosody Generation For Speech Synthesis Using Linguistics-Driven Acoustic Embedding Selection Shubhi Tyagi* (Amazon), Marco Nicolis (Amazon), Jonas Rohnke (Amazon), Thomas Drugman (Amazon), Jaime Lorenzo-Trueba (Amazon)
A smoothed GDA algorithm for the nonconvex-concave min-max problem\\ with an $\mathcal{O}\left(1/\epsilon^2\right)$ iteration complexity Jiawei Zhang (Shenzhen Research Institute of Big Data), Peijun Xiao* (University of Illinois at Urbana-Champaign), Ruoyu Sun (University of Illinois at Urbana-Champaign), Zhi-Quan Luo (Shenzhen Research Institute of Big Data)
Error Propagation Analysis of Nonnegative Tensor Train Utilized for Nonnegative Canonical Polyadic Decomposition Svetlana Kuksova* (Los Alamos National Laboratory), Erik W. Skau (Los Alamos National Laboratory), Boian S. Alexandrov ( Los Alamos National Laboratory)
Exploring aspects of similarity between spoken personal narratives by disentangling them into narrative clause types Belen Saldias* (MIT), Deb Roy (MIT)
Exponential Step Sizes for Non-Convex SGD Xiaoyu Li* (Boston University), Zhenxun Zhuang (Boston University), Franceso Orabona(Boston University)
Feature and Example Selection for Predictive Analytics of Imbalanced Traffic Crash Data Deepti Lamba* (Kansas State University), Majed Alsadhan (Kansas State University), William Hsu (Kansas State University), Eric Fitzsimmons (Kansas State University), Greg Newmark (Kansas State University)
GAN-based Generation of Synthetic Minority HEp-2 Cell Images for Autoimmune Disorders Diagnosis Krati Gupta* (IIT Mandi), Daksh Thapar (IIT Mandi), Arnav Bhavsar (IIT Mandi), and Anil K. Sao (IIT Mandi)
Group Retention when Using Machine Learning in Sequential Decision Making: the Interplay between User Dynamics and Fairness Xueru Zhang* (University of Michigan),Mohammadmahdi Khaliligarekani (University of Michigan),Cem Tekin (Bilkent University), Mingyan Liu (University of Michigan)
Efficient Cross-Modal Retrieval via Deep Binary Hashing and Quantization Yang Shi* (Rakuten Institute of Technology, Rakuten, Inc.), Young-joo Chung (Rakuten Institute of Technology, Rakuten, Inc.)
Hybrid quantum neural networks Vijayasri Iyer* (AMMAS Research Lab, Coimbatore, India), Bhargava Ganti (AMMAS Research Lab, Coimbatore, India), P K Krishnan Namboori (AMMAS Research Lab, Coimbatore, India), Sriram Iyer (Dorian Technology Solutions)
Towards Data Quality Improvement of Low-cost IoT Sensors in Environmental Monitoring Networks Using Data Fusion and Machine Learning Approach Nwamaka Okafor*, Declan Delaney (University College Dublin)
Identification of Patterns in Cystic Fibrosis Physiotherapy with Unsupervised Learning Olga Liakhovich* (Microsoft), Tempest van Schaik (Microsoft), Bianca Furtuna (Microsoft)
In-domain and Cross-domain Transferability for Damage Detection in Structural Health Monitoring Zaharah Allah Bukhsh* (Radboud University, Nijmegen), Nils Jansen 2 (Radboud University, Nijmegen) Aaqib Saeed 3 (Eindhoven University of Technology)
Mutli-Domain Image Completion for Random Missing Data with Representational Disentanglement Liyue Shen* (Stanford University), Wentao Zhu (Nvidia), Xiaosong Wang (Nvidia), Lei Xing (Stanford University), John Pauly (Stanford University), Daguang Xu (Nvidia)
Lack of representation in data – the next biggest barrier to equal healthcare and access Meghana Bhimarao* (Kaiser Permanente Division of Research)
Lagrangian-based Dynamics for Game Optimization Reyhane Askari Hemmat* (Mila, University of Montreal), Amartya Mitra (Mila, University of California Riverside), Guillaume Lajoi (Mila, University of Montreal), Ioannis Mitliagkas (Mila, University of Montreal)
Fashion meets Computer Vision: Learning to Exploit the Rich Style Space of Outfits Mariya Vasileva* (UIUC), David Forsyth (UIUC)
Memory-Efficient Pairwise Neural Networks (PairNets) for Artificial Intelligence of Things (AIoT) Applications Luna Zhang (BigBear, Inc.)
LRS-DAG: Low Resource Supervised DomainAdaptation with Generalization Across Domains Rheeya Uppaal* (University of Massachusetts, Amherst)
Long-Tail Predictions with Continuous-Output Language Models Shiran Dudy* (OHSU), Steven Bedrick (OHSU)
Hop-Hop Relation-aware Graph Neural Network Li Zhang* (University of Sheffield), Haiping Lu (University of Sheffield)
Learning-Based Strong Solutions to Forward and Inverse Problems in PDEs Leah Bar* (Tel-Aviv University, Israel), Nir Sochen (Tel-Aviv University, Israel)
Plato : Automated Mechanism to choose optimal reward strategy between short-term and long-term rewards in website Personalization Abhimanyu Mitra (Walmartlabs), Afroza Ali* (Walmartlabs), Xiaotong Suo (Walmartlabs), Kailing Wang (Walmartlabs), Kannan Achan (Walmartlabs)
Prediction of Intraseasonal Mean and Active/Break Spells for Indian Summer Monsoon Using a CNN Moumita Saha* (University of Colorado Boulder), Ravi S. Nanjundiah (Indian Institute of Tropical Meteorology), Claire Monteleoni (University of Colorado Boulder)
Prediction of Adverse Event on Drug-Drug Combination using Graph Embedding Ankita Saha* (IIT Kharagpur, India), Jayanta Mukhopadhyay ( IIT Kharagpur, India), Sudeshna Sarkar (IIT Kharagpur, India), Mahanandeeshwar Gattu (Excelra Knowledge Solutions Pvt Ltd, Hyderabad, India)
Predicting Mortality Risk in Viral and Unspecified Pneumonia to Assist Clinicians with COVID-19 ECMO Planning Helen Zhou* (Carnegie Mellon University), Cheng Cheng (Carnegie Mellon University), Zachary Lipton (Carnegie Mellon University), George Chen (Carnegie Mellon University), Jeremy Weiss (Carnegie Mellon University)
Submodular Maximization via Taylor Series Approximation Gözde Özcan* (Northeastern University), Armin Moharrer (Northeastern University), Stratis Ioannidis (Northeastern University)
Robust Sentiment Classification in Noisy User-generated Text with BERT-based model Gwenaelle Cunha Sergio* (Kyungpook National University), Minho Lee (Kyungpook National University)
Robust Large-Margin Learning in Hyperbolic Space Melanie Weber* (Princeton University), Manzil Zaheer (Google), Ankit Singh Rawat (Google), Aditya Menon (Google), Sanjiv Kumar (Google)
Review of feature importance for mental health interventions Rhythm Bhatia* (University of Eastern Finland)
Recurrent Convolutional Network based Hybrid Video Compression Aishwarya Jadhav* (VJTI, Mumbai)
WaveMix – A Multi-Task Neural Architecture for Music Performance Extraction Emily McQuillin* (Durham University), Frederick Li (Durham University).
Unsupervised Land Cover Change Detection and Interpretation from Multispectral Satellite Image Time-Series Srija Chakraboty* (NASA GSFC, Arizona State University)
Unlimited Resolution Image Generation with R2D2-GANs Marija Jegorova* (University of Edinburgh), Antti Ilari Karjalainen (Seebyte Ltd), Jose Vazquez (Seebyte Ltd), Timothy Hospedales (University of Edinburgh)
Understanding the Evolution of the #MeToo Movement Over Time Using Topic Models Sara Kangaslahti* (California Institute of Technology), Anqi Liu (California Institute of Technology), R. Michael Alvarez (California Institute of Technology), Anima Anandkumar (California Institute of Technology)
Uncertainty-Aware Search Framework for Multi-Objective Bayesian Optimization with Constraints Syrine Belakaria* (Washington State University), Aryan Deshwal (Washington State University), Jana Doppa (Washington State University)
Tracking Drift using Symbolic Data: Method & Applications Shikha Verma* (Indian Institute of Management, Ahmedabad), Arnab K. Laha (Indian Institute of Management, Ahmedabad)
Towards improved, computationally efficient feature attributions with Integrated DeepLIFT Jocelin Su* (MIT), Avanti Shrikumar (Stanford University), Anshul Kundaje (Stanford University)
Supply Chain Resilience in the Era of COVID-19: Assessment Using Machine Learning and Maritime Data Sara El Mekkaoui* (Equipe AMIPS, Ecole Mohammadia d’Ingénieurs, Mohammed V University in Rabat, Morocco), Loubna Benabbou (Université du Québec à Rimouski), Abdelaziz Berrado (Equipe AMIPS, Ecole Mohammadia d’Ingénieurs, Mohammed V University in Rabat, Morocco)
Who’s that Player? Sports Gear Segmentation for Athlete Identification Hoi-Ying Mak* (Miro AI), Wayne Fong (Miro AI), John Ho (Miro AI), King-Hei Fung (Miro AI), Cederico Martinez (Miro AI)
Explainable Artificial Intelligence Sherin Mathews* (Mcafee)

Copyright © Women in Machine Learning 2020