CIAIRI Open Day
Tue Feb 01 2022 at 10:00 am to 06:30 pm
RMIT University Storey Hall | Melbourne, VI
About this Event
Join us for showcases, workshops, demonstrations from AI in Health to Agent Oriented programming for Autonomous Systems. Learn about going beyond supervised deep learning, AI and Mixed Reality to Interactive storytelling. This day has been rescheduled to April 5th .
Dr. Azadeh Alavi - Computer Vision - Artificial Intelligence in health-How AI can be used to provide assertive tools and diagnostic guide in health care domain
Prof.Fabio Zambetta Mixed Reality AI+Mixed Reality: The future of Computer-Human Interaction The talk focuses on how a wide range of AI techniques are going to provide new types of mixed reality interfaces, leading to different kinds of applications. It will include demo e.g., showing spatial reasoning on a HoloLens and the computer vision algorithms used in it, etc.
Dr.Iman Abbasnejad Mixed Reality SCONEGAN: Semantic Contrastive learning-based Generative Adversarial Network for an end-to-end image translation SCONEGAN presents an end-to-end image translation, which is shown to be effective for learning to generate realistic and diverse scenery images. Most current image-to-image translation approaches are devised as two mappings: a translation from the source to target domain and another to represent its inverse. A short video will also be presented from the generated images using SCONEGAN.
Dr.Sebastian Rodriguez Agent based Decision Making SARL: Agent oriented programming SARL is a general-purpose agent-oriented language. SARL aims at providing the fundamental abstractions for dealing with concurrency, distribution, reactivity, autonomy and dynamic reconfiguration. This talk presents SARL’s main concepts and provides an introduction to developing MAS with SARL.
Dr.Adrian DYER Computer Vision AI lessons from the miniature brain of the bee. A challenge for AI has been operation in complex natural environments. Bees have a brain of less than 1 million neurons but can solve many complex tasks previously thought to require a human brain. This discussion will include Einstein proposed bees as a model 70 years ago, with modern psychophysics and computer modelling techniques how bees are indeed providing solutions that change our understanding of how to build AI solutions for real world applications.
Dr.Rick Evertsz Agent Based Decision Making TDF - A Methodology and Tool for Designing Intelligent Autonomous Systems TDF is a methodology and tool for diagrammatically specifying autonomous decision-making agents that can collaborate as part of a team with humans. It allows one to specify the knowledge and reasoning required to function effectively in the real world, including how to coordinate as a team. This talk will provide an overview of its capabilities and applications.
Dr.Dhirendra Singh Agent Based Decision Making Keeping us safe: AI for quantifying bushfire risk to communities and planning safe evacuations Bushfires are part of life for many Australians, and communities and authorities must consider a range of strategies to reduce the risk to lives. This talk will showcase two tools that are being developed with CSIRO's Data61 and in collaboration with the emergency services and local and state governments. First is a macro-level tool used to quickly compute community evacuation risk "hotspots" across the state for given weather conditions using up to a million bushfire spread predictions. The second is a micro-level tool that allows high fidelity community evacuation simulations to be configured and run, allowing authorities to test a range of evacuation strategies to determine a safe course of action against a specific fire threat. Together, these tools will support the emergency services in ensuring community safety in these life-threatening events.
A/Prof Jeffrey Chan Learning Technologies AI in Prescriptive Analytics This talk will discuss prescriptive analytics (decision making), particularly to do with research that combines predictions/machine learning and then optimisation to make the best choice.
Dr. Haytham Fayek Learning Technologies Beyond Supervised Deep Learning Learning representations from small, labelled datasets or large not necessarily labelled datasets for multi-task machine perception.
A/Prof Julie Porteous Learning Technologies Interactive Storytelling Application of AI in entertainment, new media and games
Dr.Minyi Li Learning Technologies Preference elicitation and learning Understanding and predicting users’ preferences play a key role in various fields, e.g., recommender systems, adaptive user interface design, general product design and brand building, autonomous systems design, etc.. However, in real-world decision problems, users’ preferences are usually very complex, i.e., they generally have multiple decision criteria and have to deal with an exponential number of choices and alternative solutions. This makes the investigation directly through preference relations/ranking over the entire solution space become ineffective and infeasible. In this talk, we will explore ways of eliciting, reasoning, and learning human preferences, as well as how we could apply the techniques in various application domains.
Dr. Huong Ha Learning Technologies Data-efficient Active Testing of Machine Learning Algorithms Machine learning models are being used extensively in many important areas, but there is no guarantee a model will always perform well or as its developers intended. Understanding the correctness of a model is crucial to prevent potential failures that may have a significant detrimental impact in critical application areas. In this talk, we propose a novel framework to efficiently test a machine learning model using only a small amount of labelled test data.
Where is it happening?RMIT University Storey Hall, 348 Swanston Street, Melbourne, Australia
Event Location & Nearby Stays: