Meet the organizers and other participants for informal socializing on Zoom.

Please join our Slack workspace, which will be active for the duration of the un-workshop and ICML.

To reach the booth please join us on Zoom and if you are asked for a password, use 559104.

Staffed Times (UTC/GMT):

8.30 to 9.00 Fannie (Recruiter)

10.40 to 11.10 Maren (Principal Data Scientist)

11.00 to 12.00 Fannie (Recruiter)

13.00 to 14.00 Marta (Recruiter)

17:40 to 18:00 Marta (Recruiter)

Drop by any of these times to chat with one of our representatives!

Special Events (UTC/GMT):

10.25 - 10.40 please join us in the WiML Expo to hear Maren Eckhoff, Principal Data Scientist, present on ‘Using AI for social and global good’.

10.40 - 11.10 please join us in the QuantumBlack booth where Maren will be available for Q&A.

More Information:

Visit our careers page for latest job postings.

If you want to get in touch, please email us at info@quantumblack.com.

To reach the booth please join our ICML sponsor page.

Staffed Times (UTC/GMT):

ICML booth in the sponsor hall is open to ICML attendees!

Special Events (UTC/GMT):

10:40 - 10:55 Maria Dimakopoulou will be giving a talk on “Slate Bandit Learning and Evaluation”.

17:00 - 19:00 she will be available at our booth to answer questions.

More Information is available at our research site - NETFLIX RESEARCH.

To reach the booth please join our ICML sponsor page.

Staffed Times (UTC/GMT):

The booth will be staffed from 13.00 to 15.00

More Information is available at our research site - IBM Research.

To reach the booth please join our ICML sponsor page.

Staffed Times (UTC/GMT):

Please visit the Apple booth in ICML's virtual sponsor expo to chat with members of our machine learning teams and recruiting teams.

Special Events (UTC/GMT):

Join the sponsor expo at the WiML un-workshop to hear Lizi Ottens from our CoreML team talk about machine learning at Apple.

Lizi will also be available to chat 1:1 at our ICML booth on Tuesday, July 14th 18:00 - 19:00, and during our “Q&A with Apple” session on Tuesday, July 14th 19:30-20:00.

More Information: Visit the AI/ML team page on Apple Jobs.

To reach the booth please join our ICML sponsor page.

Staffed Times (UTC/GMT):

Visit our booth by any of the below times to chat with a DeepMinder:

07:30 - 08:00 10:00 - 14:00 16:45 - 17:00

Special Events (UTC/GMT):

15:40 – 15:55 Meire Fortunato will be introducing the work of DeepMind during our Sponsor Expo.

She will be available at the booth to answer questions right after her talk ends.

To reach the booth please join our Google Meet.

Staffed Times (UTC/GMT):

We are excited to connect with you at WiML! Visit our booth by any of the below times to chat with a Googler:

08:00 - 09:00 14:00 - 15:10 17:00 - 18:00

Special Events (UTC/GMT):

We also welcome you to hear from Jennifer Wei during the WiML Expo, who will be giving a talk entitled “Smell: Learning Generalizable Perceptual Representations of Small Molecules.”

The talk will be streamed from 15:10 - 15:25 and then Jennifer Wei will give a live Q&A in our booth from 15:25 - 15:55.

More Information: Learn more about the research we’re presenting, including accepted papers, workshops, and tutorials at our ICML Virtual booth.

To reach the booth please join our ICML sponsor page.

Special events (UTC/GMT):

15:55 - 16:10 Kalesha Bullard will be giving a talk on “Learning to Communicate Nonverbally for Embodied Agent Populations”.

She will be available at the booth to answer questions right after her talk ends.

More information: For more information, please visit our careers page.

Please watch the pre-recorded talk on SlidesLive.

Please watch the pre-recorded talk on SlidesLive.

Please watch the pre-recorded talk on SlidesLive.

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.

Please post & upvote questions using Sli.do.

Leaders: Gagana Coolga, Stuti Gupta

Facilitators: Akash Smaran

If you are asked for a Zoom password, please use 931226.

Leaders: Mary-Anne Hartley, Danielle Belgrave

Facilitators: Berthine Nyunga

If you are asked for a Zoom password, please use 353535.

Leaders: Aneta Neumann, Frehiwot Girmay

Facilitators: Aparna Akula, Tina Raissi

If you are asked for a Zoom password, please use 566888.

Leaders: Surangika Ranathunga, Rishemjit Kaur, Annie En-Shiun Lee

Facilitators: Marjana Skenduli, Mehreen Alam

If you are asked for a Zoom password, please use 312711.

Please watch the pre-recorded talk on SlidesLive.

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.

Please post & upvote questions using Sli.do.

Leaders: Ines Krissaane, Samrudhdhi Rangrej

Facilitators: Laya Rafiee

If you are asked for a Zoom password, please use 392590.

Leaders: Adepeju Oshisanya, Allison Gardner, Aylin Cakiroglu, Simone Larsson

Facilitators: Celine Lature

If you are asked for a Zoom password, please use 156371.

Please watch the pre-recorded talk on SlidesLive.

Please watch the pre-recorded talk on SlidesLive.

Please watch the pre-recorded talk on SlidesLive.

Meet the organizers and other participants for informal socializing on Zoom.

Please watch the pre-recorded talk on SlidesLive.

Please watch the pre-recorded talk on SlidesLive.

Please watch the pre-recorded talk on SlidesLive.

Please watch the pre-recorded talk on SlidesLive.

Please watch the pre-recorded talk on SlidesLive.

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.

Please post & upvote questions using Sli.do.

Leaders: Olga Liakhovich, Tempest van Schaik, Summer Elasady, Bianca Furtuna

Facilitators: Katie Claveau

If you are asked for a Zoom password, please use 021583.

Leaders: Ghazal Fazelnia, Zahra Nazari, Mozhgan Saeidi

Facilitators: Krystal Maughan, Sneha Srinivasan

If you are asked for a Zoom password, please use 309398.

Leaders: Tai-Danae Bradley

Facilitators: Melanie Weber

If you are asked for a Zoom password, please use 549805.

Leaders: Natalie Stanley, Huda Nassar, Ina Stelzer

Facilitators: Jolene Ranek

If you are asked for a Zoom password, please use 683278.

Leaders: Fatemehsadat Mireshghallah, Srishti Yadav, Mary Anne Smart

Facilitators: Jin (Alice) Qixuan

If you are asked for a Zoom password, please use 996705.

Leaders: Priya Donti, Sasha Luccioni

Facilitators: David Rolnick

If you are asked for a Zoom password, please use 77918249.

Leaders: Jennifer Hobbs, Saba Dadsetan, Naira Hovakimyan Facilitators: Tania Lorido Botran, Lori Liu

If you are asked for a Zoom password, please use 252034.

Leaders: Reyhane Askari Hemmat, Alexia Jolicoeur-Martineau, Laya Rafiee

Facilitators: Xing Han

If you are asked for a Zoom password, please use 893180.

Leaders: Weiwei Zong, Manju Liu

Facilitators: Zhen Sun

If you are asked for a Zoom password, please use 344525.

Leaders: Chhavi Yadav, Irina Higgins, Jovana Mitrović

Facilitators: Laure Delisle, Niveditha Kalavakonda

If you are asked for a Zoom password, please use 07132020.

Please watch the pre-recorded talk on SlidesLive.

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.

Please post & upvote questions using Sli.do.

Leaders: Anoush Najarian, Ishaani, Aleshia Hayes

Facilitators: Sindhuja Parimalarangan, Louvere Walker-Hannon

If you are asked for a Zoom password, please use 565501.

Leaders: Swetasudha Panda, Emily Black, Xueru Zhang

Facilitators: Shikha Bordia

If you are asked for a Zoom password, please use 21647420.

Leaders: Vidhi Jain, Simin Liu

Facilitators: Ganesh Iyer

If you are asked for a Zoom password, please use 890211.

Leaders: Celestine Mendler-Dünner, Tijana Zrnic

Facilitators: Juan Carlos Perdomo

If you are asked for a Zoom password, please use 13072020.

Leaders: Sergul Aydore, Haleh Akrami

Facilitators: Berna Kabadayi

If you are asked for a Zoom password, please use 221530.

Leaders: Elvisha Dhamala, Meenakshi Khosla

Facilitators: Carmen Khoo

If you are asked for a Zoom password, please use 112211.

Leaders: Vaidheeswaran Archana, Sherin Mathews, Yashika Sharma

Facilitators: Zahra Vaseqi

If you are asked for a Zoom password, please use 179681.

Leaders: Khimya Khetarpal, Rose E. Wang, Feryal Behbahani

Facilitators: Arundhati Banerjee

If you are asked for a Zoom password, please use 925172.

Leaders: Polina Kirichenko, Melanie F. Pradier, Weiwei Pan

Facilitators: Ana-Denisa Secuiu

If you are asked for a Zoom password, please use 294409.

Leaders: Belen Saldias, Safinah Ali

Facilitators: Tamara Covacevich, Clare Liu

If you are asked for a Zoom password, please use 123789.

Please post & upvote questions using Sli.do.

Please watch the pre-recorded talk on SlidesLive.

This live poster session will be held in Gather.town.

See here for the listing of posters that will be presented.

Please note:

ICML registration required to enter.

Entry is first-come-first-serve.

If you are not able to enter, please check back again later, as people will be coming in and out of the Gather.town space, just like any in-person space.

Besides this live poster session in Gather.town, each WiML poster has a Slack channel in the WiML Slack that is active for the duration of ICML, and certain posters also pre-recorded 5-min talks on SlidesLive.

Please fill-in the form.