First Australasian Symposium on Artificial Intelligence for the Environment (AI4Environment)
Workshop Delivery: 9 December 10am – 3pm (AWST/UTC+8) (Hybrid Mode: face-to-face and online)
- A Layer-Based Decentralised Interoperability Model for IoT Edge Devices; Tanzima Azad, Jarrod Trevathan, Ma Hakim Newton and Abdul Sattar
- Building Metadata Inference Using a Transducer Based Language Model; David Waterworth, Subbu Sethuvenkatraman and Quan Z. Sheng
- Machine Learning-based Classification of Birds through Birdsong; Richard Sinnott and Yueying Chang
- An operational framework to automatically evaluate the quality of weather observations from third-party stations; Quanxi Shao, Ming Li, Joel Dabrowski, Shuvo Bakar, Ashfaqur Rahman, Andrea Powell and Brent Henderson
Title: Artificial intelligence based digital transformation for decision support in agricultural
Climatic and environmental changes play major roles in various adversaries in the agricultural processes, e.g., disease, pests, weeds, heat and cold stress. These can significantly impact the values of crops in terms of yields and quality. It is, therefore, imperative for the farmers to monitor the crops at various growth stages. Such methods were traditionally manual or semi-automatic, thereby making the process labour-intensive, error-prone and often too late for intervention. Digital technologies offer better data acquisition and analysis tools that are further empowered by AI technologies. As such the formation of such adversaries can be detected very early which can support decision-making about mitigation strategies. In this talk, we are going to cover background and state-of-the-art in weed detection and recognition, heat and cold stress monitoring in crop plants, and disease and pest detection and classification. We will discuss how imaging technologies have been transforming data acquisition and the advancements in artificial intelligence techniques are used in analyzing those data.
A/Prof Ferdous Sohelreceived a PhD degree from Monash University, Australia. He received BSc Engineering degree in Computer Science and Engineering from Bangladesh University of Engineering and Technology (BUET). He is currently an Associate Professor in Information Technology at Murdoch University, Australia. Prior to joining Murdoch University, he was a Research Assistant Professor/Research Fellow at the School of Computer Science and Software Engineering, The University of Western Australia.
Ferdous Sohel’s broad field of research is “Information and Computing Sciences” with “Artificial Intelligence and Computer Vision” being his main expertise. He investigates the areas of advanced image/video/ data analysis and machine learning techniques with applications in visual scene understanding, acoustic scene analysis, medical/bio-medical applications, digital agricultural technologies, environmental sustainability measures, remote sensing, and surveillance and monitoring. He leads the Digital Agriculture research theme of Murdoch University. He has published well over 190 peer reviewed international journal and conference articles.
Prof Sohel is an Associate Editor of IEEE Transactions on Multimedia (2020-2023) and IEEE Signal Processing Letters (2021-2022). He was a Technical Program Chair of DICTA2019 and DICTA2021, Organising Secretary of APCC2017, and tutorial Chair of PSIVT 2019. He was a co-presenter of a tutorial at CVPR2015. He served as an International Program Committee member of many conferences. He is a member of the Australian Computer Society and a senior member of the IEEE. He served the IEEE Western Australia Section as the Vice-Chair (in 2016 and 2017).
Paper submission: 06 November Notification: 14 November
Workshop Website: https://easychair.org/cfp/ai4environment22
Submission Link: https://easychair.org/conferences/?conf=ai4environment
Quanxi Shao, CSIRO
Ming Li, CSIRO
Subbu Sethuvenkatraman, CSIRO
Prof Abdul Sattar, Griffith University
As Artificial Intelligence (AI) drives changes to enhance productivity and prolong lifespans, a dual concern is the environmental sustainability of scarce planetary resources. Effective management of environmental assets and reducing our impact on sensitive ecosystems is becoming a critical issue. AI has the potential to accelerate global efforts to protect the environment and conserve resources through renewable energy systems, driving energy emission reductions, CO2 removal, developing greener transportation networks, monitoring deforestation, and planning for extreme weather conditions. This workshop aims to explore the emerging applications of AI as it relates to environmental conservation and management.
List of Topics
Papers are being solicited on (but not limited to) the following topics:
- Air Quality Monitoring and Forecasting
- Climate Adaptation
- Disaster Resilience Planning (Bush Fires, Flooding, Droughts, Tsunami, Storms)
- E-Waste/Plastic Recycling/Up-Cycling (Green IT and the Circular Economy)
- Energy Sustainability (Distributed Energy Resource Portfolios)
- Environmental Financial Planning, Insurance and Economics
- Home Automation
- Internet of Things (IoT)/Internet of Everything (IoE)/AI of Things (AIoT) Environmental Sensing
- Smart Cities, Infrastructure and Buildings (Urban Monitoring and Optimisation)
- Sustainable Agriculture/Aquaculture
- Sustainable Production/Manufacturing
- Transport Management and Optimisation
- Water Quality Monitoring and Forecasting
- Weather Monitoring and Forecasting
The First Workshop on Toxic Language Detection (TLD)
Leaderboard Challenges Due : 24 November, 2022
Announcement of Winners: 28 November, 2022
Camera-ready Abstract Due: 2 December, 2022
TLD Workshop: 6 December, 2022 11AM – 4PM (AWST / UTC+8)
Workshop Website : https://tld2022.github.io/
The growth of online content presents challenges to detect toxic language using Natural Language Processing (NLP). Fast detection of such language can enable mitigation strategies to be acted to thwart toxic situations. In this workshop, we use toxic language as a broader concept that can be overlapped with other terms such as abusive language, hate speech, or offensive language. The goal of the workshop is to bring together researchers and practitioners to engage in a discussion about identifying and detecting such toxic language.
The workshop consists of
- a series of invited talks by reputed members of the NLP community on toxicity detection from academia related to the topics below;
- a shared task with leaderboard challenges for CONDA in-game toxicity dataset; and
- a presentation and ceremony for the leaderboard challenges.
Topics of interest include, but are not limited to:
- NLP Models for Toxic Language Detection across different domains (e.g. social media, news comments, Wikipedia, and in-game chat).
- Multi-modal Models for Toxic Language Detection (e.g. A combination of using text, audio, and image).
- Development of Low Resource Language for Toxicity Detection.
- Multi-lingual Dataset and Models for Toxic Language Detection.
- Applications of Toxic Language Detection.
- Human-Computer Interaction for Toxic Language Detection System.
- Bias in Toxic Language Detection Systems.
To promote the research and practice, a shared task for CONDA Toxicity Detection Challenge will be held in November 2022, and two winning team will be awarded a cash prize. This competition will provide a good testbed for participants to develop better toxicity detection systems.
Traditional toxicity detection models have focused on the single utterance level without deeper understanding of context. The CONDA dataset is to detect in-game toxic language, enabling joint intent classification and slot filling analysis, which is a core task in Natural Language Understanding (NLU). The dataset consists of 45K utterances from 12K conversations from the chat logs of 1.9K completed the Defense of the Ancients 2 (Dota 2) matches. Dota 2 is a multiplayer online game where teams of five players attempt to destroy their opponents’ ancient structure. In this challenge, participants are to implement a model for Joint Slot and Intent Classification and to evaluate their results for toxicity language detection task via leaderboard.
Federated learning in Australasia: Towards large-scale AI system with privacy-preserving
9 December, 10am – 3pm (AWST/UTC+8)
In contrast to traditional centralized machine learning techniques, federated learning (FL) is a learning technique that trains machine learning models across multiple decentralized edge devices holding local data samples without exchanging them. FL enables multiple end-users to build a secure and robust machine learning model without sharing data, addressing critical issues such as data privacy, data security, data access rights, and fairness. It has gained interest from various entities in both academia and industry. This workshop aims to bring together academic researchers and industry AI practitioners to address cutting-edge works contributing to federated machine learning research and its application to real-world applications, especially in Australasian area.
We intend to create a forum to communicate challenging problems and address open issues in this research area, and federated learning related innovative applications in Australasian. We also would like to invite PhD students in Australasian to exchange ideas about their published Federated Learning papers at top AI conferences recent.
The authors can submit their unpublished papers. The authors can also submit the papers for sharing their recent works published papers at top conferences (NeurIPS, ICML, ICLR, CVPR, ICCV, ACL, AAAI, IJCAI, etc.) with a slide/poster. A prerecorded video demonstrating the research is also encouraged for application. All accepted and presented papers will be published on the website. We are not seeking to publish proceedings, thus the authors can also submit their papers to other conferences and journals with formal publication. The workshop organizers will review them and select high-quality papers to be shared and discussed in the workshop