This is an international workshop on weakly supervised learning. The main goal of this workshop is to discuss challenging research topics in weakly supervised learning areas such as semi-supervised learning, positive-unlabeled learning, label noise learning, partial label learning, and self-supervised learning, as well as the foster collaborations among universities and institutes.

Program

Venue & Date

Venue: OpenSpace of RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Tokyo, Japan. Please click here for more access details of RIKEN AIP.

Date: November 23 09:30--17:40, and November 24 09:30--16:00 (Asia/Tokyo time (UTC+9)).

URL for Online Attending: https://c5dc59ed978213830355fc8978.doorkeeper.jp/events/164877

Schedule

The workshop will use Asia/Tokyo time (UTC+9) for scheduling, and it will be combined with invited talks and contributed talks.

Topics

Overview

Machine learning should not be accessible only to those who can pay. Specifically, modern machine learning is migrating to the era of complex models (e.g., deep neural networks), which require a plethora of well-annotated data. Giant companies have enough money to collect well-annotated data. However, for startups or non-profit organizations, such data is barely acquirable due to the cost of labeling data or the intrinsic scarcity in the given domain. These practical issues motivate us to research and pay attention to weakly supervised learning (WSL), since WSL does not require such a huge amount of annotated data. We define WSL as the collection of machine learning problem settings and algorithms that share the same goals as supervised learning but can only access to less supervised information than supervised learning. In this workshop, we discuss both theoretical and applied aspects of WSL.

This workshop is a series of our previous workshops at ACML 2019, SDM 2020, ACML 2020, IJCAI 2021, and ACML 2021. Our particular technical emphasis at this workshop is incomplete supervision, inexact supervision, inaccurate supervision, cross-domain supervision, imperfect demonstration, and weak adversarial supervision. Meanwhile, this workshop will also focus on WSL for Science and Social Good, such as WSL for COVID-19, WSL for healthcare, WSL for climate change, WSL for remote sensing, and new public WSL datasets regarding the scientific scenarios.

Topics of Interest

WSL workshop includes (but not limited to) the following topics:

• Algorithms and theories of incomplete supervision, e.g., semi-supervised learning, active learning, and positive-unlabeled learning;
• Algorithms and theories of inexact supervision, e.g., multi-instance learning, complementary learning, and open-set learning;
• Algorithms and theories of inaccurate supervision, e.g., crowdsourced learning and label-noise learning;
• Algorithms and theories of cross-domain supervision, e.g., zero-/one-/few-shot learning, transferable learning, and multi-task learning;
• Algorithms and theories of imperfect demonstration, e.g., inverse reinforcement learning and imitation learning with non-expert demonstrations;
• Algorithms and theories of adversarial weakly-supervised learning, e.g., adversarial semi-supervised learning and adversarial label-noisy learning;
• Algorithms and theories of self-supervision, e.g., contrastive learning and autoencoder learning;
• Broad applications of weakly supervised learning in the field of computer science, such as weakly supervised object detection (computer vision), weakly supervised sequence modeling (natural language processing), and weakly supervised cross-media retrieval (information retrieval).
• WSL for science and social good, such as WSL for COVID-19, WSL for healthcare, WSL for climate change, and WSL for remote sensing, meanwhile, new public datasets regarding the above WSL research directions (new focus).

Further Descriptions

The focus of this workshop is six types of weak supervision: incomplete supervision, inexact supervision, inaccurate supervision, cross-domain supervision, imperfect demonstration, and weak adversarial supervision, which are briefly introduced below.

• Incomplete supervision considers a subset of training data given with ground-truth labels while the other data remain unlabeled, such as semi-supervised learning and positive-unlabeled learning.
• Inexact supervision considers the situation where some supervision information is given but not as exacted as desired, i.e., only coarse-grained labels are available. For example, if we are considering to classify every pixel of an image, rather than the image itself, then ImageNet becomes a benchmark with inexact supervision. Besides, multi-instance learning belongs to inexact supervision, where we do not exactly know which instance in the bag corresponds to the given ground-truth label.
• Inaccurate supervision considers the situation where the supervision information is not always the ground-truth, such as label-noise learning.
• Cross-domain supervision considers the situation where the supervision information is scarce or even non-existent in the current domain but can be possibly derived from other domains. Examples of cross-domain supervision appear in zero-/one-/few-shot learning, where external knowledge from other domains is usually used to overcome the problem of too few or even no supervision in the original domain.
• Imperfect demonstration considers the situation for inverse reinforcement learning and imitation learning, where the agent learns with imperfect or non-expert demonstrations. For example, AlphaGo learns a policy from a sequence of states and actions (expert demonstration). Even if an expert player wins a game, it is not guaranteed that every action in the sequence is optimal.
• Weak adversarial supervision considers the situation where weak supervision meets adversarial robustness. Since machine learning models are increasingly deployed in real-world applications, their security attracts more and more attention from both academia and industry. Therefore, many robust learning algorithms aim to prevent various evasion attacks, e.g., adversarial attacks, privacy attacks, model stealing attacks, and so on. However, almost all those robust algorithms (against evasion attacks) implicitly assume the strong supervision signals (no noisy labels in the training data), which hardly meets the requirements in practice. Therefore, when we develop evasion-robust algorithms, it is very practical/urgent to consider the supervision signals are imperfect.
• Self-supervision considers the unsupervised situation, and it pre-trains a generic feature representation by autonomously building the pseudo supervision (e.g., the similarity contrast and sample reconstruction) from the raw data, the learned representation can be applied in various downstream tasks such as classification, retrieval, and clustering.

Meanwhile, this workshop will continue discussing broad applications of weakly supervised learning in the field of computer science, such as weakly supervised object detection (computer vision), weakly supervised sequence modeling (natural language processing), and weakly supervised cross-media retrieval (information retrieval).

Organizers

Program Chairs

Shuo Chen, RIKEN, Japan.

Yang Liu, University of California, Santa Cruz, USA.

Feng Liu, The University of Melbourne, Australia.

Bo Han, Hong Kong Baptist University, Hong Kong SAR, China.

Tongliang Liu, The University of Sydney, Australia.

Gang Niu, RIKEN, Japan.

Masashi Sugiyama, RIKEN / The University of Tokyo, Japan.

Advisory Board (alphabetical order by last name)

Chen Gong, Nanjing University of Science and Technology, China.

Mingming Gong, The University of Melbourne, Australia.

Yu-Feng Li, Nanjing University, China.

Miao Xu, The University of Queensland, Australia.

Quanming Yao, Tsinghua University / 4Paradigm Inc., China.

Jingfeng Zhang, The University of Auckland, New Zealand.

Min-Ling Zhang, Southeast University, China.

Previous Workshops

ACML2022 WSL Workshop, Online.

ACML2021 WSL Workshop, Online.

IJCAI2021 WSRL Workshop, Online.

ACML2020 WSRL Workshop, Online.

SDM2020 WSUL Workshop, Ohio, United States.

ACML2019 WSL Workshop, Nagoya, Japan.