The last decade saw an enormous boost in the field of computational topology: methods and concepts from algebraic and differential topology, formerly confined to the realm of pure mathematics, have demonstrated their utility in numerous areas such as computational biology, personalised medicine, materials science, and time-dependent data analysis, to name a few.

The newly-emerging domain comprising topology-based techniques is often referred to as topological data analysis (TDA). Next to their applications in the aforementioned areas, TDA methods have also proven to be effective in supporting, enhancing, and augmenting both classical machine learning and deep learning models.

We believe that it is time to bring together theorists and practitioners in a creative environment to discuss the goals beyond the currently-known bounds of TDA. We want to start a conversation between experts, non-experts, and users of TDA methods to debate the next steps the field should take. We also want to disseminate methods to a broader audience and demonstrate how easy the integration of topological concepts into existing methods can be.

Important links:

- Rocket.Chat (for asking questions)
- Slack (for asking questions)

This one day workshop focuses on privacy preserving techniques for machine learning and disclosure in large scale data analysis, both in the distributed and centralized settings, and on scenarios that highlight the importance and need for these techniques (e.g., via privacy attacks). There is growing interest from the Machine Learning (ML) community in leveraging cryptographic techniques such as Multi-Party Computation (MPC) and Homomorphic Encryption (HE) for privacy preserving training and inference, as well as Differential Privacy (DP) for disclosure. Simultaneously, the systems security and cryptography community has proposed various secure frameworks for ML. We encourage both theory and application-oriented submissions exploring a range of approaches listed below. Additionally, given the tension between the adoption of machine learning technologies and ethical, technical and regulatory issues about privacy, as highlighted during the COVID-19 pandemic, we invite submissions for the special track on this topic.

How to join the virtual workshop: The 2020 Workshop on Meta-Learning will be a series of streamed pre-recorded talks + live question-and-answer (Q&A) periods, and poster sessions on Gather.Town. You can participate by:
* Accessing the livestream on our [ protected link dropped ] 2;
* MetaLearn 2020 Rocket.Chat!
* Entering panel discussion questions in this sli.do!


Focus of the workshop: Recent years have seen rapid progress in meta-learning methods, which transfer knowledge across tasks and domains to learn new tasks more efficiently, optimize the learning process itself, and even generate new learning methods from scratch. Meta-learning can be seen as the logical conclusion of the arc that machine learning has undergone in the last decade, from learning classifiers and policies over hand-crafted features, to learning representations over which classifiers and policies operate, and finally to learning algorithms that themselves acquire representations, classifiers, and policies. Meta-learning methods are also of substantial practical interest. For instance, they have been shown to yield new state-of-the-art automated machine learning algorithms and architectures, and have substantially improved few-shot learning systems. Moreover, the ability to improve one’s own learning capabilities through experience can also be viewed as a hallmark of intelligent beings, and there are strong connections with work on human learning in cognitive science and reward learning in neuroscience.

Climate change is one of the greatest problems society has ever faced, with increasingly severe consequences for humanity as natural disasters multiply, sea levels rise, and ecosystems falter. Since climate change is a complex issue, action takes many forms, from designing smart electric grids to tracking greenhouse gas emissions through satellite imagery. While no silver bullet, machine learning can be an invaluable tool in fighting climate change via a wide array of applications and techniques. These applications require algorithmic innovations in machine learning and close collaboration with diverse fields and practitioners. This workshop is intended as a forum for those in the machine learning community who wish to help tackle climate change. Building on our past workshops on this topic, this workshop aims to especially emphasize the pipeline to impact, through conversations about machine learning with decision-makers and other global leaders in implementing climate change strategies. The all-virtual format of NeurIPS 2020 provides a special opportunity to foster cross-pollination between researchers in machine learning and experts in complementary fields.

Optimization lies at the heart of many machine learning algorithms and enjoys great interest in our community. Indeed, this intimate relation of optimization with ML is the key motivation for the OPT series of workshops.

Looking back over the past decade, a strong trend is apparent: The intersection of OPT and ML has grown to the point that now cutting-edge advances in optimization often arise from the ML community. The distinctive feature of optimization within ML is its departure from textbook approaches, in particular, its focus on a different set of goals driven by "big-data, nonconvexity, and high-dimensions," where both theory and implementation are crucial.

We wish to use OPT 2020 as a platform to foster discussion, discovery, and dissemination of the state-of-the-art in optimization as relevant to machine learning. And well beyond that: as a platform to identify new directions and challenges that will drive future research, and continue to build the OPT+ML joint research community.

Invited Speakers
Volkan Cevher (EPFL)
Michael Friedlander (UBC)
Donald Goldfarb (Columbia)
Andreas Krause (ETH, Zurich)
Suvrit Sra (MIT)
Rachel Ward (UT Austin)
Ashia Wilson (MSR)
Tong Zhang (HKUST)

Instructions
Please join us in gather.town for all breaks and poster sessions (Click "Open Link" on any break or poster session).

To see all submitted paper and posters, go to the "opt-ml website" at the top of the page.

Use RocketChat or Zoom link (top of page) if you want to ask the speaker a direct question during the Live Q&A and Contributed Talks.

This workshop will explore how advances in machine learning could be applied to improve educational outcomes.

Such an exploration is timely given: the growth of online learning platforms, which have the potential to serve as testbeds and data sources; a growing pool of CS talent hungry to apply their skills towards social impact; and the chaotic shift to online learning globally during COVID-19, and the many gaps it has exposed.

The opportunities for machine learning in education are substantial, from uses of NLP to power automated feedback for the substantial amounts of student work that currently gets no review, to advances in voice recognition diagnosing errors by early readers.

Similar to the rise of computational biology, recognizing and realizing these opportunities will require a community of researchers and practitioners that are bilingual: technically adept at the cutting-edge advances in machine learning, and conversant in most pressing challenges and opportunities in education.

With representation from senior representatives from industry, academia, government, and education, this workshop is a step in that community-building process, with a focus on three things:
1. identifying what learning platforms are of a size and instrumentation that the ML community can leverage,
2. building a community of experts bringing rigorous theoretical and methodological insights across academia, industry, and education, to facilitate combinatorial innovation,
3. scoping potential Kaggle competitions and “ImageNets for Education,” where benchmark datasets fine tuned to an education goal can fuel goal-driven algorithmic innovation.

In addition to bringing speakers across verticals and issue areas, the talks and small group conversations in this workshop will be designed for a diverse audience--from researchers, to industry professionals, to teachers and students. This interdisciplinary approach promises to generate new connections, high-potential partnerships, and inspire novel applications for machine learning in education.

​This workshop is not the first Machine Learning for Education workshop; there has been several (ml4ed.cc), and the existence of these others speaks to recognition of the the obvious importance that ML will have for education moving forward!

Recent years have seen a surge in research at the intersection of differential geometry and deep learning, including techniques for stochastic optimization on curved spaces (e.g., hyperbolic or spherical manifolds), learning embeddings for non-Euclidean data, and generative modeling on Riemannian manifolds. Insights from differential geometry have led to new state of the art approaches to modeling complex real world data, such as graphs with hierarchical structure, 3D medical data, and meshes.
Thus, it is of critical importance to understand, from a geometric lens, the natural invariances, equivariances, and symmetries that reside within data.

In order to support the burgeoning interest of differential geometry in deep learning, the primary goal for this workshop is to facilitate community building and to work towards the identification of key challenges in comparison with regular deep learning, along with techniques to overcome these challenges. With many new researchers beginning projects in this area, we hope to bring them together to consolidate this fast-growing area into a healthy and vibrant subfield. In particular, we aim to strongly promote novel and exciting applications of differential geometry for deep learning with an emphasis on bridging theory to practice which is reflected in our choices of invited speakers, which include both machine learning practitioners and researchers who are primarily geometers.

Quantum tensor networks in machine learning (QTNML) are envisioned to have great potential to advance AI technologies. Quantum machine learning promises quantum advantages (potentially exponential speedups in training, quadratic speedup in convergence, etc.) over classical machine learning, while tensor networks provide powerful simulations of quantum machine learning algorithms on classical computers. As a rapidly growing interdisciplinary area, QTNML may serve as an amplifier for computational intelligence, a transformer for machine learning innovations, and a propeller for AI industrialization.

Tensor networks, a contracted network of factor tensors, have arisen independently in several areas of science and engineering. Such networks appear in the description of physical processes and an accompanying collection of numerical techniques have elevated the use of quantum tensor networks into a variational model of machine learning. Underlying these algorithms is the compression of high-dimensional data needed to represent quantum states of matter. These compression techniques have recently proven ripe to apply to many traditional problems faced in deep learning. Quantum tensor networks have shown significant power in compactly representing deep neural networks, and efficient training and theoretical understanding of deep neural networks. More potential QTNML technologies are rapidly emerging, such as approximating probability functions, and probabilistic graphical models. However, the topic of QTNML is relatively young and many open problems are still to be explored.

Quantum algorithms are typically described by quantum circuits (quantum computational networks). These networks are indeed a class of tensor networks, creating an evident interplay between classical tensor network contraction algorithms and executing tensor contractions on quantum processors. The modern field of quantum enhanced machine learning has started to utilize several tools from tensor network theory to create new quantum models of machine learning and to better understand existing ones.

The interplay between tensor networks, machine learning and quantum algorithms is rich. Indeed, this interplay is based not just on numerical methods but on the equivalence of tensor networks to various quantum circuits, rapidly developing algorithms from the mathematics and physics communities for optimizing and transforming tensor networks, and connections to low-rank methods for learning. A merger of tensor network algorithms with state-of-the-art approaches in deep learning is now taking place. A new community is forming, which this workshop aims to foster.

The application of machine learning to healthcare is often characterised by the development of cutting-edge technology aiming to improve patient outcomes. By developing sophisticated models on high-quality datasets we hope to better diagnose, forecast, and otherwise characterise the health of individuals. At the same time, when we build tools which aim to assist highly-specialised caregivers, we limit the benefit of machine learning to only those who can access such care. The fragility of healthcare access both globally and locally prompts us to ask, “How can machine learning be used to help enable healthcare for all?” - the theme of the 2020 ML4H workshop.

Participants at the workshop will be exposed to new questions in machine learning for healthcare, and be prompted to reflect on how their work sits within larger healthcare systems. Given the growing community of researchers in machine learning for health, the workshop will provide an opportunity to discuss common challenges, share expertise, and potentially spark new research directions. By drawing in experts from adjacent disciplines such as public health, fairness, epidemiology, and clinical practice, we aim to further strengthen the interdisciplinarity of machine learning for health.

See our workshop for more information: https://ml4health.github.io/

This workshop is designed to bring together trainees and experts in machine learning with those in the very forefront of biological research today for this purpose. Our full-day workshop will advance the joint project of the CS and biology communities with the goal of "Learning Meaningful Representations of Life" (LMRL), emphasizing interpretable representation learning of structure and principle. As last year, the workshop will be oriented around four layers of biological abstraction: molecule, cell, synthetic biology, and phenotypes.

Mapping structural molecular detail to organismal phenotype and function; predicting emergent effects of human genetic variation; and designing novel interventions including prevention, diagnostics, therapeutics, and the development of new synthetic biotechnologies for causal investigations are just some of the challenges that hinge on appropriate formal structures to make them accessible to the broadest possible community of computer scientists, statisticians, and their tools.

Classical machine learning research has been focused largely on models, optimizers, and computational challenges. As technical progress and hardware advancements ease these challenges, practitioners are now finding that the limitations and faults of their models are the result of their datasets. This is particularly true of deep networks, which often rely on huge datasets that are too large and unwieldy for domain experts to curate them by hand. This workshop addresses issues in the following areas: data harvesting, dealing with the challenges and opportunities involved in creating and labeling massive datasets; data security, dealing with protecting datasets against risks of poisoning and backdoor attacks; policy, security, and privacy, dealing with the social, ethical, and regulatory issues involved in collecting large datasets, especially with regards to privacy; and data bias, related to the potential of biased datasets to result in biased models that harm members of certain groups. Dates and details can be found at securedata.lol

Conversational interaction systems such as Amazon Alexa, Google Assistant, Apple Siri, and Microsoft Cortana have become very popular over the recent years. Such systems have allowed users to interact with a wide variety of content on the web through a conversational interface. Research challenges such as the Dialogue System Technology Challenges, Dialogue Dodecathlon, Amazon Alexa Prize and the Vision and Language Navigation task have continued to inspire research in conversational AI. These challenges have brought together researchers from different communities such as speech recognition, spoken language understanding, reinforcement learning, language generation, and multi-modal question answering.
Unlike other popular NLP tasks, dialogue frequently has humans in the loop, whether it is for evaluation, active learning or online reward estimation. Through this workshop we aim to bring together researchers from academia and industry to discuss the challenges and opportunities in such human in the loop setups. We hope that this sparks interesting discussions about conversational agents, interactive systems, and how we can use humans most effectively when building such setups. We will highlight areas such as human evaluation setups, reliability in human evaluation, human in the loop training, interactive learning and user modeling. We also highly encourage non-English based dialogue systems in these areas.
The one-day workshop will include talks from senior technical leaders and researchers to share insights associated with evaluating dialogue systems. We also plan on having oral presentations and poster sessions on works related to the topic of the workshop. Finally we will end the workshop with an interactive panel of speakers. As an outcome we expect the participants from the NeurIPS community to walk away with better understanding of human in the loop dialogue modeling as well as key areas of research in this field. Additionally we would like to see discussions around the unification of human evaluation setups in some way.

This workshop will consist of live QA sessions. Therefore, in order to get the most out of the workshop, it is recommended that you watch all the prerecorded talks before the workshop day. Additionally we have put Reserved blocks of time as an opportunity to watch the pre-recorded talks before the Q/A.

Machine learning research has benefited considerably from the adoption of standardised public benchmarks. In this workshop proposal, we do not argue against the importance of these benchmarks, but rather against the current incentive system and its heavy reliance upon performance as a proxy for scientific progress. The status quo incentivises researchers to “beat the state of the art”, potentially at the expense of deep scientific understanding and rigorous experimental design. Since typically only positive results are rewarded, the negative results inevitably encountered during research are often omitted, allowing many other groups to unknowingly and wastefully repeat the same negative findings. Pre-registration is a publishing and reviewing model that aims to address these issues by changing the incentive system. A pre-registered paper is a regular paper that is submitted for peer-review without any experimental results, describing instead an experimental protocol to be followed after the paper is accepted. This implies that it is important for the authors to make compelling arguments from theory or past published evidence. As for reviewers, they must assess these arguments together with the quality of the experimental design, rather than comparing numeric results. In this workshop, we propose to conduct a full pilot study in pre-registration for machine learning. It follows a successful small-scale trial of pre-registration in computer vision and is more broadly inspired by the success of pre-registration in the life sciences.

“Differentiable programs” are parameterized programs that allow themselves to be rewritten by gradient-based optimization. They are ubiquitous in modern-day machine learning. Recently, explicitly encoding our knowledge of the rules of the world in the form of differentiable programs has become more popular. In particular, differentiable realizations of well-studied processes such as physics, rendering, projective geometry, optimization to name a few, have enabled the design of several novel learning techniques. For example, many approaches have been proposed for unsupervised learning of depth estimation from unlabeled videos. Differentiable 3D reconstruction pipelines have demonstrated the potential for task-driven representation learning. A number of differentiable rendering approaches have been shown to enable single-view 3D reconstruction and other inverse graphics tasks (without requiring any form of 3D supervision). Differentiable physics simulators are being built to perform physical parameter estimation from video or for model-predictive control. While these advances have largely occurred in isolation, recent efforts have attempted to bridge the gap between the aforementioned areas. Narrowing the gaps between these otherwise isolated disciplines holds tremendous potential to yield new research directions and solve long-standing problems, particularly in understanding and reasoning about the 3D world.

Hence, we propose the “first workshop on differentiable computer vision, graphics, and physics in machine learning” with the aim of:
1. Narrowing the gap and fostering synergies between the computer vision, graphics, physics, and machine learning communities
2. Debating the promise and perils of differentiable methods, and identifying challenges that need to be overcome
3. Raising awareness about these techniques to the larger ML community
4. Discussing the broader impact of such techniques, and any ethical implications thereof.

Causality is a fundamental notion in science and engineering, and one of the fundamental problems in the field is how to find the causal structure or the underlying causal model. For instance, one focus of this workshop is on causal discovery, i.e., how can we discover causal structure over a set of variables from observational data with automated procedures? Another area of interest is how a causal perspective may help understand and solve advanced machine learning problems.

Recent years have seen impressive progress in theoretical and algorithmic developments of causal discovery from various types of data (e.g., from i.i.d. data, under distribution shifts or in nonstationary settings, under latent confounding or selection bias, or with missing data), as well as in practical applications (such as in neuroscience, climate, biology, and epidemiology). However, many practical issues, including confounding, the large scale of the data, the presence of measurement error, and complex causal mechanisms, are still to be properly addressed, to achieve reliable causal discovery in practice.

Moreover, causality-inspired machine learning (in the context of transfer learning, reinforcement learning, deep learning, etc.) leverages ideas from causality to improve generalization, robustness, interpretability, and sample efficiency and is attracting more and more interest in Machine Learning (ML) and Artificial Intelligence. Despite the benefit of the causal view in transfer learning and reinforcement learning, some tasks in ML, such as dealing with adversarial attacks and learning disentangled representations, are closely related to the causal view but are currently underexplored, and cross-disciplinary efforts may facilitate the anticipated progress.

This workshop aims to provide a forum for discussion for researchers and practitioners in machine learning, statistics, healthcare, and other disciplines to share their recent research in causal discovery and to explore the possibility of interdisciplinary collaboration. We also particularly encourage real applications, such as in neuroscience, biology, and climate science, of causal discovery methods.

***
After each keynote, there will be 5 minutes for a live Q&A. You may post your questions in Rocket.Chat before or during the keynote time. The poster session and the virtual coffee break will be on Gather.Town. There is no Q&A for orals and spotlight talks, but all papers will attend the poster session and you can interact with authors there. More details will come soon.

There is a trend in the machine learning community to adopt self-supervised approaches to pre-train deep networks. Self-supervised learning utilizes proxy supervised learning tasks, for example, distinguishing parts of the input signal from distractors, or generating masked input segments conditioned on the unmasked ones, to obtain training data from unlabeled corpora. These approaches make it possible to use a tremendous amount of unlabeled data on the web to train large networks and solve complicated tasks. ELMo, BERT, and GPT in NLP are famous examples in this direction. Recently self-supervised approaches for speech and audio processing are also gaining attention. These approaches combine methods for utilizing no or partial labels, unpaired text and audio data, contextual text and video supervision, and signals from user interactions. Although the research direction of self-supervised learning is active in speech and audio processing, current works are limited to several problems such as automatic speech recognition, speaker identification, and speech translation, partially due to the diversity of modeling in various speech and audio processing problems. There is still much unexplored territory in the research direction for self-supervised learning.

This workshop will bring concentrated discussions on self-supervision for the field of speech and audio processing via several invited talks, oral and poster sessions with high-quality papers, and a panel of leading researchers from academia and industry. Alongside research work on new self-supervised methods, data, applications, and results, this workshop will call for novel work on understanding, analyzing, and comparing different self-supervision approaches for speech and audio processing. The workshop aims to:
- Review existing and inspire new self-supervised methods and results,
- Motivate the application of self-supervision approaches to more speech and audio processing problems in academia and industry, and encourage discussion amongst experts and practitioners from the two realms,
- Encourage works on studying methods for understanding learned representations, comparing different self-supervision methods and comparing self-supervision to other self-training as well as transfer learning methods that low-resource speech and audio processing have long utilized,
- Facilitate communication within the field of speech and audio processing (e.g., people who attend conferences such as INTERSPEECH and ICASSP) as well as between the field and the whole machine learning community for sharing knowledge, ideas, and data, and encourage future collaboration to inspire innovation in the field and the whole community.

For the eighth edition of the CiML (Challenges in Machine Learning) workshop at NeurIPS, our goals are to: 1) Increase diversity in the participant community in order to increase quality of model predictions; 2) Identify and share best practices in building AI capability in vulnerable communities; 3) Celebrate pioneers from these communities who are modeling lifelong learning, curiosity and courage in learning how to use ML to address critical problems in their communities.

The workshop will provide concrete recommendations to the ML community on designing and implementing competitions that are more accessible to a broader public, and more effective in building long-term AI/ML capability.

The workshop will feature keynote speakers from ML, behavioral science and gender and development, interspersed with small group discussions around best practices in implementing ML competitions. We will invite submissions of 2-page extended abstracts on topics relating to machine learning competitions, with a special focus on methods of creating diverse datasets, strategies for addressing behavioral barriers to participation in ML competitions from underrepresented communities, and strategies for measuring the long-term impact of participation in an ML competition.

It has become increasingly clear in the recent years that AI research, far from producing neutral tools, has been concentrating power in the hands of governments and companies and away from marginalized communities. Unfortunately, NeurIPS has lacked a venue explicitly dedicated to understanding and addressing the root of these problems. As Black feminist scholar Angela Davis famously said, "Radical simply means grasping things at the root." Resistance AI exposes the root problem of AI to be how technology is used to rearrange power in the world. AI researchers engaged in Resistance AI both resist AI that centralizes power into the hands of the few and dream up and build human/AI systems that put power in the hands of the people. This workshop will enable AI researchers in general, researchers engaged in Resistance AI, and marginalized communities in particular to reflect on AI-fueled inequity and co-create tactics for how to address this issue in our own work.

Logistics:
We will use the main/webinar Zoom + livestream for most events, with interactive events taking place on a separate auxiliary/breakout Zoom or gather.town. Please see our workshop site for details: https://sites.google.com/view/resistance-ai-neurips-20/schedule
See also our welcome doc here for further detail, including community guidelines and where each activity can be found: http://bit.ly/rai-welcome

Machine learning methods have had great success in learning complex representations that enable them to make predictions about unobserved data. Physical sciences span problems and challenges at all scales in the universe: from finding exoplanets in trillions of sky pixels, to finding machine learning inspired solutions to the quantum many-body problem, to detecting anomalies in event streams from the Large Hadron Collider. Tackling a number of associated data-intensive tasks including, but not limited to, segmentation, 3D computer vision, sequence modeling, causal reasoning, and efficient probabilistic inference are critical for furthering scientific discovery. In addition to using machine learning models for scientific discovery, the ability to interpret what a model has learned is receiving an increasing amount of attention.

In this targeted workshop, we would like to bring together computer scientists, mathematicians and physical scientists who are interested in applying machine learning to various outstanding physical problems, in particular in inverse problems and approximating physical processes; understanding what the learned model really represents; and connecting tools and insights from physical sciences to the study of machine learning models. In particular, the workshop invites researchers to contribute papers that demonstrate cutting-edge progress in the application of machine learning techniques to real-world problems in physical sciences, and using physical insights to understand what the learned model means.

By bringing together machine learning researchers and physical scientists who apply machine learning, we expect to strengthen the interdisciplinary dialogue, introduce exciting new open problems to the broader community, and stimulate production of new approaches to solving open problems in sciences. Invited talks from leading individuals in both communities will cover the state-of-the-art techniques and set the stage for this workshop.

—

Gather Town link: [ protected link dropped ]

Learning-based methods, and in particular deep neural networks, have emerged as highly successful and universal tools for image and signal recovery and restoration. They achieve state-of-the-art results on tasks ranging from image denoising, image compression, and image reconstruction from few and noisy measurements. They are starting to be used in important imaging technologies, for example in GEs newest computational tomography scanners and in the newest generation of the iPhone.

The field has a range of theoretical and practical questions that remain unanswered. In particular, learning and neural network-based approaches often lack the guarantees of traditional physics-based methods. Further, while superior on average, learning-based methods can make drastic reconstruction errors, such as hallucinating a tumor in an MRI reconstruction or turning a pixelated picture of Obama into a white male.

This virtual workshop aims at bringing together theoreticians and practitioners in order to chart out recent advances and discuss new directions in deep neural network-based approaches for solving inverse problems in the imaging sciences and beyond. NeurIPS, with its visibility and attendance by experts in machine learning, offers the ideal frame for this exchange of ideas. We will use this virtual format to make this topic accessible to a broader audience than the in-person meeting is able to as described below.

In the proposed workshop, we aim to discuss the challenges and opportunities for machine learning research in the context of physical systems. This discussion involves the presentation of recent methods and the experiences made during the deployment on real-world platforms. Such deployment requires a significant degree of generalization. Namely, the real world is vastly more complex and diverse compared to fixed curated datasets and simulations. Deployed machine learning models must scale to this complexity, be able to adapt to novel situations, and recover from mistakes. Moreover, the workshop aims to strengthen further the ties between the robotics and machine learning communities by discussing how their respective recent directions result in new challenges, requirements, and opportunities for future research.

Following the success of previous robot learning workshops at NeurIPS, the goal of this workshop is to bring together a diverse set of scientists at various stages of their careers and foster interdisciplinary communication and discussion.
In contrast to the previous robot learning workshops which focused on applications in robotics for machine learning, this workshop extends the discussion on how real-world applications within the context of robotics can trigger various impactful directions for the development of machine learning. For a more engaging workshop, we encourage each of our senior presenters to share their presentations with a PhD student or postdoctoral researcher from their lab. Additionally, all our presenters - invited and contributed - are asked to add a ``dirty laundry’’ slide, describing the limitations and shortcomings of their work. We expect this will aid further discussion in poster and panel sessions in addition to helping junior researchers avoid similar roadblocks along their path.

Welcome to the NeurIPS 2020 Workshop on Machine Learning for Autonomous Driving!

Autonomous vehicles (AVs) offer a rich source of high-impact research problems for the machine learning (ML) community; including perception, state estimation, probabilistic modeling, time series forecasting, gesture recognition, robustness guarantees, real-time constraints, user-machine communication, multi-agent planning, and intelligent infrastructure. Further, the interaction between ML subfields towards a common goal of autonomous driving can catalyze interesting inter-field discussions that spark new avenues of research, which this workshop aims to promote. As an application of ML, autonomous driving has the potential to greatly improve society by reducing road accidents, giving independence to those unable to drive, and even inspiring younger generations with tangible examples of ML-based technology clearly visible on local streets.

All are welcome to submit and/or attend! This will be the 5th NeurIPS workshop in this series. Previous workshops in 2016, 2017, 2018 and 2019 enjoyed wide participation from both academia and industry.

Recent advances in deep reinforcement learning and robotics have enabled agents to achieve superhuman performance on a variety of challenging games and learn complex manipulation tasks. While these results are very promising, several open problems remain. In order to function in real-world environments, learned policies must be both robust to input perturbations and be able to rapidly generalize or adapt to novel situations. Moreover, to collaborate and live with humans in these environments, the goals and actions of embodied agents must be interpretable and compatible with human representations of knowledge. Hence, it is natural to consider how humans so successfully perceive, learn, and plan to build agents that are equally successful at solving real world tasks.
There is much evidence to suggest that objects are a core level of abstraction at which humans perceive and understand the world [8]. Objects have the potential to provide a compact, casual, robust, and generalizable representation of the world. Recently, there have been many advancements in scene representation, allowing scenes to be represented by their constituent objects, rather than at the level of pixels. While these works have shown promising results, there is still a lack of agreement on how to best represent objects, how to learn object representations, and how best to leverage them in agent training.
In this workshop we seek to build a consensus on what object representations should be by engaging with researchers from developmental psychology and by defining concrete tasks and capabilities that agents building on top of such abstract representations of the world should succeed at. We will discuss how object representations may be learned through invited presenters with expertise both in unsupervised and supervised object representation learning methods. Finally, we will host conversations and research on new frontiers in object learning.

Despite the obvious advantages, automation driven by machine learning and artificial intelligence carries pitfalls for the lives of millions of people: disappearance of many well-established mass professions and consumption of labeled data that are produced by humans managed by out of time approach with full-time office work and pre-planned task types. Crowdsourcing methodology can be considered as an effective way to overcome these issues since it provides freedom for task executors in terms of place, time and which task type they want to work on. However, many potential participants of crowdsourcing processes hesitate to use this technology due to a series of doubts (that have not been removed during the past decade).

This workshop brings together people studying research questions on

(a) quality and effectiveness in remote crowd work;
(b) fairness and quality of life at work, tackling issues such as fair task assignment, fair work conditions, and on providing opportunities for growth; and
(c) economic mechanisms that incentivize quality and effectiveness for requester while maintaining a high level of quality and fairness for crowd performers (also known as workers).

Because quality, fairness and opportunities for crowd workers are central to our workshop, we will invite a diverse group of crowd workers from a global public crowdsourcing platform to our panel-led discussion.

Workshop web site: [ protected link dropped ]

The financial services industry has unique needs for fairness when adopting artificial intelligence and machine learning (AI/ML). First and foremost, there are strong ethical reasons to ensure that models used for activities such as credit decisioning and lending are fair and unbiased, or that machine reliance does not cause humans to miss critical pieces of data. Then there are the regulatory requirements to actually prove that the models are unbiased and that they do not discriminate against certain groups.

Emerging techniques such as algorithmic credit scoring introduce new challenges. Traditionally financial institutions have relied on a consumer’s past credit performance and transaction data to make lending decisions. But, with the emergence of algorithmic credit scoring, lenders also use alternate data such as those gleaned from social media and this immediately raises questions around systemic biases inherent in models used to understand customer behavior.

We also need to play careful attention to ways in which AI can not only be de-biased, but also how it can play an active role in making financial services more accessible to those historically shut out due to prejudice and other social injustices.

The aim of this workshop is to bring together researchers from different disciplines to discuss fair AI in financial services. For the first time, four major banks have come together to organize this workshop along with researchers from two universities as well as SEC and FINRA (Financial Industry Regulatory Authority). Our confirmed invited speakers come with different backgrounds including AI, law and cultural anthropology, and we hope that this will offer an engaging forum with diversity of thought to discuss the fairness aspects of AI in financial services. We are also planning a panel discussion on systemic bias and its impact on financial outcomes of different customer segments, and how AI can help.

First session for the competition program at NeurIPS2020.

Machine learning competitions have grown in popularity and impact over the last decade, emerging as an effective means to advance the state of the art by posing well-structured, relevant, and challenging problems to the community at large. Motivated by a reward or merely the satisfaction of seeing their machine learning algorithm reach the top of a leaderboard, practitioners innovate, improve, and tune their approach before evaluating on a held-out dataset or environment. The competition track of NeurIPS has matured in 2020, its fourth year, with a considerable increase in both the number of challenges and the diversity of domains and topics. A total of 16 competitions are featured this year as part of the track, with 8 competitions associated to each of the two days. The list of competitions that ar part of the program are available here:

https://neurips.cc/Conferences/2020/CompetitionTrack

The exponential growth of AI research has led to several papers floating on arxiv, making it difficult to review existing literature. Despite the huge demand, the proportion of survey & analyses papers published is very low due to reasons like lack of a venue and incentives. Our Workshop, ML-RSA provides a platform and incentivizes writing such types of papers. It meets the need of taking a step back, looking at the sub-field as a whole and evaluating actual progress. We will accept 3 types of papers: broad survey papers, meta-analyses, and retrospectives. Survey papers will mention and cluster different types of approaches, provide pros and cons, highlight good source code implementations, applications and emphasize impactful literature. We expect this type of paper to provide a detailed investigation of the techniques and link together themes across multiple works. The main aim of these will be to organize techniques and lower the barrier to entry for newcomers. Meta-Analyses, on the other hand, are forward-looking, aimed at providing critical insights on the current state-of-affairs of a sub-field and propose new directions based on them. These are expected to be more than just an ablation study -- though an empirical analysis is encouraged as it can provide for a stronger narrative. Ideally, they will seek to showcase trends that are not possible to be seen when looking at individual papers. Finally, retrospectives seek to provide further insights ex post by the authors of a paper: these could be technical, insights into the research process, or other helpful information that isn’t apparent from the original work.

In recent years, the use of deep neural networks as function approximators has enabled researchers to extend reinforcement learning techniques to solve increasingly complex control tasks. The emerging field of deep reinforcement learning has led to remarkable empirical results in rich and varied domains like robotics, strategy games, and multiagent interactions. This workshop will bring together researchers working at the intersection of deep learning and reinforcement learning, and it will help interested researchers outside of the field gain a high-level view about the current state of the art and potential directions for future contributions.

Machine learning (ML) has seen a tremendous amount of recent success and has been applied in a variety of applications. However, it comes with several drawbacks, such as the need for large amounts of training data and the lack of explainability and verifiability of the results. In many domains, there is structured knowledge (e.g., from electronic health records, laws, clinical guidelines, or common sense knowledge) which can be leveraged for reasoning in an informed way (i.e., including the information encoded in the knowledge representation itself) in order to obtain high quality answers. Symbolic approaches for knowledge representation and reasoning (KRR) are less prominent today - mainly due to their lack of scalability - but their strength lies in the verifiable and interpretable reasoning that can be accomplished. The KR2ML workshop aims at the intersection of these two subfields of AI. It will shine a light on the synergies that (could/should) exist between KRR and ML, and will initiate a discussion about the key challenges in the field.

Deep neural network models have shown remarkable performance in tasks such as visual object recognition, speech recognition, and autonomous robot control. We have seen continuous improvements throughout the years which have led to these models surpassing human performance in a variety of tasks such as image classification, video games, and board games. However, the performance of deep learning models heavily relies on a massive amount of data, which requires huge time and effort to collect and label them.

Recently, to overcome these weaknesses and limitations, attention has shifted towards machine learning paradigms such as semi-supervised learning, incremental learning, and meta-learning which aim to be more data-efficient. However, these learning models still require a huge amount of data to achieve high performance on real-world problems. There has been only a few achievement or breakthrough, especially in terms of the ability to grasp abstract concepts and to generalize problems.

In contrast, human babies gradually make sense of the environment through their experiences, a process known as learning by doing, without a large amount of labeled data. They actively engage with their surroundings and explore the world through their own interactions. They gradually acquire the abstract concept of objects and develop the ability to generalize problems. Thus, if we understand how a baby's mind develops, we can imitate those learning processes in machines and thereby solve previously unsolved problems such as domain generalization and overcoming the stability-plasticity dilemma. In this workshop, we explore how these learning mechanisms can help us build human-level intelligence in machines.

In this interdisciplinary workshop, we bring together eminent researchers in Computer Science, Cognitive Science, Psychology, Brain Science, Developmental Robotics and various other related fields to discuss the below questions on babies vs. machines.

■ How far is the state-of-the-art machine intelligence from babies?
■ How does a baby learn from their own interactions and experiences?
■ What sort of insights can we acquire from the baby's mind?
■ How can those insights help us build smart machines with baby-like intelligence?
■ How can machines learn from babies to do better?
■ How can these machines further contribute to solving the real-world problems?

We will invite selected experts in the related fields to give insightful talks. We will also encourage interdisciplinary contributions from researchers in the above topics. Hence, we expect this workshop to be a good starting point for participants in various fields to discuss theoretical fundamentals, open problems, and major directions of further development in an exciting new area.

www.mlforeconomicpolicy.com
mlforeconomicpolicy.neurips2020@gmail.com

The goal of this workshop is to inspire and engage a broad interdisciplinary audience, including computer scientists, economists, and social scientists, around topics at the exciting intersection of economics, public policy, and machine learning. We feel that machine learning offers enormous potential to transform our understanding of economics, economic decision making, and public policy, and yet its adoption by economists and social scientists remains nascent.

We want to use the workshop to expose some of the critical socio-economic issues that stand to benefit from applying machine learning, expose underexplored economic datasets and simulations, and identify machine learning research directions that would have significant positive socio-economic impact. In effect, we aim to accelerate the use of machine learning to rapidly develop, test, and deploy fair and equitable economic policies that are grounded in representative data.

For example, we would like to explore questions around whether machine learning can be used to help with the development of effective economic policy, to understand economic behavior through granular, economic data sets, to automate economic transactions for individuals, and how we can build rich and faithful simulations of economic systems with strategic agents. We would like to develop economic policies and mechanisms that target socio-economic issues including diversity and fair representation in economic outcomes, economic equality, and improving economic opportunity. In particular, we want to highlight both the opportunities as well as the barriers to adoption of ML in economics.