UPDATE:

Come to the Meetup information session being held on the 6th:

https://www.meetup.com/dsai-syd/events/277266713/

Also, check out the game in action thanks to the GUI built in Python here:

https://medium.com/dsai/your-mission-if-you-choose-to-accept-it-5d08328ead66

1. INTRO

In conjunction with Australian Department of Defence, DSAi is proud to announce we are hosting a Gameathon to help generate ideas around the use of AI in the field of wargaming. The term “gameathon” is a derivation of wargaming and ideathons which are a popular way to come up with new sorts of ideas with data and technology.

Defence Science & Technology Group (DSTG) has a wargame called “Disaster of the JOADIA Islands” (detailed below) and has set up 2 challenges which cover Game Design and AI Assistance. Contestants can register as individuals or as teams and create submissions for either or both challenges. DSTG and DSAi will judge the entries and award prize.

alternative

2. BACKGROUND

Disaster of the JOADIA Islands is a turn-based wargame that models a Joint Task force assigned with the goal of rescuing civilians in a fictitious humanitarian aid disaster relief (HADR) scenario.

Playing a wargame has the potential to help human decision makers with a range of decisions related to real world problem scenarios. JOADIA is the computerised version of the original board game rule sets for the purpose of conducting research into AI and machine learning agents as a means of providing enhanced decision making. JOADIA is one of our sandbox toy platforms for investigating this potential for future emerging AI decision support concepts.

Rules of the game

3. CHALLENGES

The challenges described below have been considered from two perspectives: 1) wargaming models for real world problems and 2) the role of artificial intelligence in the wargaming process. While they are posed as separate challenges, there is a relationship between the two and we encourage participants to think about the possible ways the challenges complement one other. You may wish to consider a response for the AI challenge from the perspective of a game design challenge question and vice versa. We look forward to seeing what you come up with!
alternative

Game Design

Design new game systems and rule sets for the original game to explore concepts applicable to a HADR scenario. Some ideas that you might like to consider include:

  • Actions

    Our representation of an adversarial force is currently a militia with a limited set of actions.
    Can you come up other actions that red force could do that would disrupt blue force’s mission? How would you model them?

  • Better Immersion

    How can we design the game for better immersion?
    For example, can the game influence player perceptions on the loss of civilian lives?

  • Game Rules

    Think of an AI or autonomy-related technology that Disaster of the JOADIA Islands could model within the game rules.
    For example, what if we had autonomous drones that could deliver supplies quickly over rough terrain? How would you model this? What if the red force had AI capabilities? How could they counter blue more effectively?

  • Use Cases

    What are some modelling additional use cases that may be applicable in an HADR scenario?
    E.g working with non-government organisations (NGOs), disease models, social media.

AI Assistance

We are interested in exploring all the different possible ways that modern AI techniques could be used as part of wargaming for enhanced decision support. Some of the questions related to this that we are interested in exploring include:

  • AI uses

    What are the most effective ways that we can use AI to help players think about alternate strategies available to them?

  • Players

    In what ways can we use AI to help players understand the trade-offs between the different blue force unit capabilities? For example, what is the worth of one infantry unit versus one medic unit?

  • Analysing decisions

    How can we use AI to help in analysing the decisions made by human players? Are there any novel visualisations or maybe “intelligent” playback mechanisms that would help most effectively in this area?

  • Game Rules HADR

    How could AI be used to modify the game rules to better explore the space of HADR decision-making?

alternative

4. TIMEFRAMES

The following dates show the timeframe for submissions and judging (specific dates will be confirmed soon):

  • 15th March 2021

    Entries open to the public.

  • March 2021

    Frequently Asked Questions to be answered in post on this events page.

  • 11th April 2021

    Entries close.

  • 19th April 2021

    Final presentations to judges.

  • Late April 2021

    Final judging completed and winners announced.

5. SUBMISSIONS

  • If interested, entrants must fill in the following Google Form


  • Details of what is required are shown in the Terms and Conditions.


  • Additionally, entrants must provide answers to the above challenge in either Word, PDF, PPT as attachments to their submissions.


  • Separate documents must be provided for challenge 1 and 2. This can be in document, video or even HTML formats.


  • These must be submitted to the following DSAi email address: events@dsai.org.au


PANEL

Paul is a cofounder and director of Core Intelligence, a full stack machine learning consultancy. Outside of this, Paul is a director of the non for profit - the Data science & Ai association of Australia, where we fulfill our mission of connecting the Machine Learning and Data Science communities of Australia through education, events and our own dedicated coworking space. Paul has extensive partnerships with Microsoft, Amazon, Google & has even hosted events with the likes of Tesla. Paul is host to the podcast "The Humans of Ai" which is available on all podcast providers and is one of the co-creators of StarAi, a critically acclaimed advanced machine learning course

Paul Conyngham – DSAi

Martin works for the Defence Science and Technology Group (DSTG) in the Red Teaming & Wargaming discipline. His career at DSTG has covered a diverse range of activities including tactical military planning, complex systems science, human performance experimentation, agent based modelling, data analysis, artificial intelligence for information fusion and wargaming. He is currently involved with DST’s Modelling Complex Warfighting SRI looking at opportunities for artificial intelligence in the wargaming process. Martin’s professional qualifications include a Bachelor’s in Computer Science (Honours first class) as well as a Master’s in Defence Science (Cognitive science). His personal interests include indie game development, artificial intelligence and self-improvement

Martin Wong, DSTG

Marcin Nowina-Krowicki's professional interests lie in the research and development of novel User Interfaces to support situational awareness and decision making. Over the past two decades his work has focused on the development of Embodied Conversational Agents (Virtual Humans), immersive multi-modal geospatial displays and multimedia systems to support automated explanation and data storytelling. Recent work has included the development of a multimedia narrative explanation capability in a Command & Control (C2) system allowing a single human supervisor to manage a team of over a dozen heterogeneous autonomous vehicles.
His current work at DST in the Artificial Intelligence for Decision Analytics group explores using Multimedia Narrative, Narrative Visualisation and Provenance as a mechanism for providing eXplainable AI (XAI) in complex systems.

Marcin Nowina Krowicki, DSTG

Following a 12 year career in mainframe operations with the (then) Department of Social Security, Greg read mathematics and philosophy at the University of Tasmania, gaining a BSc (hons) first class. He then studied with the Logic and Computation group at the Australian National University, gaining a PhD in computer science. In 2009 Greg joined DST where he has worked on many things including knowledge representation, domain specific languages, plan/goal/intent recognition, human autonomy teaming and reinforcement learning. His current role is in the Artificial Intelligence for Decision Analytics group.

Greg O’Keefe, DSTG

Steve Nouri is a data science leader who has evolved the way people look at AI and innovations. An entrepreneur, investor, author, academic and technical manager by profession, he aims to inspire people through the latest technology trends and projects and empowers prospective data scientists through high-quality education and learning materials.
He is a member of the Forbes Technology Council, ICT Professional of the Year Gold Disruptor, and an accomplished influencer on LinkedIn. From humble beginnings as a software engineer and startup founder, he has worked in different IT roles before joining the Australian Computer Society as the Head of Data Science and AI, and as an AI evangelist has spoken at countless international conferences run by IBM, JP Morgan, CSIRO and AWS.
He currently lives in Sydney, Australia, and spends his time sharing technical knowledge with prospective students and advising public policy towards a more sustainable and beneficial understanding of AI and deep tech.

Steve Nouri, Australian Computer Society

Angela Kim, Head of Analytics and AI at Teacher's Health Fund and Women in AI Education Ambassador for Australia, and global head.
Angela is an AI Professional specialised in Ethics, Bias, Diversity & Inclusion in AI and recipient of Top 10 Analytics Leaders 2020 From Institute of Analytics Professional Australia and also sits on the Founding Editorial Board of Springer's new & timely AI and Ethics Journal.

Angela Kim, Women in AI

Christopher Manning is an Operations Research specialist with the Land Capability Analysis branch of Joint and Operations Analysis Division of Defence Science and Technology Group. He is currently the discipline leader for Wargaming and manages a team developing and using table-top and computer assisted wargames as well as combat simulations and other models to identify, understand and explore future Land, and more recently Joint, force options, concepts and requirements.
Chris graduated from The University of Adelaide with a Bachelor in Engineering (with Honours) majoring in Electrical and Electronic Engineering in 2001. He started work at the Defence Science and Technology Group in 2002 and completed a Masters of Science in Military Vehicle Technology at Cranfield University during a Defence Science Fellowship to the Defence Academy of the UK in 2010. He has been involved with Land 400 and, Land 907 (Australian Armoured Fighting Vehicle modernisation programs), development of the Defence Logistics Transformation Plan. Chris has also been a part of the DST Group Operations Support Centre. He is also a currently serving Officer in the Australian Army Reserve in the Australian Army’s Royal Australian Armoured Corp with over 20 years’ experience in a range of command, training, development and recruiting positions.
Chris has been married to his wife, Alison, for 16 years and they have a six year old daughter, Lillian. He hopes to be able to return to playing Ultimate Frisbee soon.

Christopher Manning, Defence Science and Technology Group

PRIZES


There are 1st, 2nd and 3rd place prizes for each challenge in the sums of $1,000, $500 and $250.
If teams are created then the prizemoney is divided amongst each member.

FIRST
1st
  • $1,000


SECOND
2nd
  • $500


THIRD
3rd
  • $250


These prizes are for each challenge and the total prizepool is $3,500.

SCREENS