COMPETITIONS

CC1: AI methods for determining ocean ecosystems from space: Combining genomic information, microscopic and satellite imagery

July 29 (Friday), 11:15-12:30, Schubert 1

The ocean is the Earth’s principal climate regulator and the main responsible for sequestering carbon dioxide (CO2). This makes it our main defense against climate change, but climate change itself is destroying the healing capacity of the ocean. Algae and, in particular, plankton, play a fundamental role in this, as they are able to remove CO2. Therefore, the mitigating capacity of an ecosystem can be established based on the presence of particular types of plankton. However, the health of the larger areas of the ocean can only be determined through large-scale measurements such as satellite imagery. The challenge focuses on the remote identification via satellite imagery of high-potential ecosystems. This would allow large tracts of the ocean to be analyzed in a way that allows scientists and decision makers to understand how the ocean evolves over time and could be used to create policies for protecting high-value parts of the ocean. Alternatively, we propose to study the use of marker species, such as whales, which can be identified and their presence implies the existence of others. This is an opportunity to attract the AI/ML community to this type of scientifically challenging and high-impact problem. For this we will make available to participants curated georeferenced datasets of plankton images, genomic data and satellite images and provide mentorship during the period of the challenge.

Timeline of the competition:

March 18th, 2022: Call for participation dissemination
April 15th-22nd, 2022: Online mini-workshop and mentorship for (prospective) participants
May 20th, 2022: Contributions submission (code + short paper)
May 31st, 2022: Decision notifications and open reviewer forum
June 10th, 2022: Final version submission (code + short paper)
June 17th, 2022: Challenge results available online

Competition url: https://oceania.inria.cl/#ai-ocean-challenge-2022

CC2: Mahjong AI Competition

July 29 (Friday), 11:15-12:30, Schubert 1

The task of this competition is to develop a computer program to play National Standard Mahjong. Mahjong is a four-player imperfect-information game, which is challenging to solve due to its complex scoring rule and rich hidden information. Participants can upload their agents to our online platform, Botzone, which has an ELO ranking system and a tournament system to evaluate their performance. The first Mahjong competition was held at IJCAI 2020, and algorithms using deep learning turned out to perform well in this game. However, the top agents submitted by participants still cannot beat the best humans. The long-term goal of this competition is to build an intelligent agent that can beat professional players in this game, and we will continue to hold it until this game is solved.

Timeline of the competition:

March 14th, 2022: Call for participation
May 22th, 2022: Submission deadline of elimination round
May 29th, 2022: Notification to finalists
July 3rd, 2022: Submission deadline of final round
July 10th, 2022: Notification of final result

Competition url: https://botzone.org.cn/static/gamecontest2022a.html

CC3: Automated Negotiating Agents Competition (ANAC)

July 26 (Tuesday), 09:00-12:30, Schubert 1

The Automated Negotiating Agent Competition (ANAC) is an international tournament that has been running since 2010 to bring together researchers from the negotiation community, which provides a unique benchmark for evaluating practical negotiation strategies in multi-issue domains. This year, we have three different negotiation research challenges: Automated Negotiation League (GeniusWeb framework), Human-Agent Negotiation (IAGO framework), and Supply Chain Management (NegMas framework). We expect innovative and novel agent strategies will be developed, and the submitted ANAC 2022 agents will serve as a negotiating agent repository to the negotiation community. The researchers can develop novel negotiating agents and evaluate their agents by comparing their performance with the performance of the ANAC 2022 agents.

Timeline of the competition:

June 10th, 2022: Submission deadline for the competition
July 1st, 2022 (at the latest): Notification of the finalists
The winner will be announced during the ANAC session which will be held in conjunction with IJCAI-ECAI 2022.

Competition url: https://web.tuat.ac.jp/~katfuji/ANAC2022/index.html

CC4: 1st Learning and Mining with Noisy Labels Challenge

July 29 (Friday), 11:15-12:30, Schubert 1

The goal of the challenge is to prepare the weakly supervised learning community to shift from synthetic label noise benchmarks to tasks with real human-annotated data. The competition will organize a series of sub-competitions that focus on a variety of learning tasks on self-collected and human-annotated CIFAR datasets, which we name as CIFAR-N.

Timeline of the challenge:

April 4th, 2022: Call for participation
April 23rd-30th, 2022: Testing access to the published dataset. Online Q&A (Slack channel, Zoom sessions)
June 15th, 2022: Deadline for participation registration (not required to submit the abstract)
June 22nd, 2022: Deadline for submission of code repository and the report
July 10th, 2022: Winners will be announced officially, and are contacted privately
July 23rd-29th, 2022: Winners will be invited to present during a dedicated session at IJCAI 2022

Challenge url: http://competition.noisylabels.com/

CC5: Geometry Friends Cooperative Game AI Competition

July 28 (Thursday), 09:00-12:30 and 14:00-17:30, Schubert 1

The goal of the competition is to build AI agents for a 2 player collaborative physics-based puzzle platformer game (Geometry Friends). Each agent controls a different character (circle or rectangle) with distinct characteristics. Their goal is to collaborate in order to collect all diamonds in a set of levels as fast as possible. The competition raises interesting problems for AI agents. For example, to successfully solve a GF level the AI players need to: (1) deal with coordination at different layers: from motion control (e.g. achieving perfect timing) to level resolution (e.g. devising shared plans); deal with limited actuation situated in a simulated physics environment (with gravity and friction); (3) solve platform (skill) based puzzles, which involves discovering the proper order to collect the diamonds and identifying the points where collaboration is needed; and (4) do all the above in real-time.

Timeline of the competition:

April, 2022: The competition is open for submissions
July 22nd, 2022: Deadline for submissions
The results will be announced during IJCAI-ECAI 2022

Competition url: https://gaips.inesc-id.pt/geometryfriends/

CC6: The Neural MMO Challenge

July 26 (Tuesday), 14:00-17:30, Schubert 1

The Neural MMO Challenge aims to develop your agents to survive and thrive together in large open worlds. The challenge focuses on decision problems of cooperation and competition by exploring, foraging, and fighting with adversaries and competitors. Participants need to implement a multi-agent algorithm to beat built-in AIs in the PvE track and compete with other algorithms in the PvP track.

Timeline of the competition:

April 14th, 2022: Challenge launches and PvE track starts
May 5th, 2022: PvP track starts
April-July 2022: The organizers will run the PvE and PvP tracks of the competition and collect the results
July (during IJCAI-ECAI 2022): Announcement of the results, award ceremony

Competition url: https://www.aicrowd.com/challenges/ijcai-2022-the-neural-mmo-challenge

CC7: AI Olympics Competition

July 29 (Friday), 11:15-12:30, Schubert 1

Developing one agent to participate in variant Olympics games is an interesting and challenging task. In this competition, participants are going to use various methods in artificial intelligence to develop agents playing different Olympics games. The games include football, running, table hockey, wrestling, etc. Agents interact with environments by receiving observations from environments and sending out actions to environments. Reinforcement learning methods can be used to solve the problems. In this competition, we provide one track using the integrated Olympics environment, which combines different Olympics games. The evaluation of the competition is going to use this environment to select different games to test the submitted agents. Different games have different rules and goals. Thus, the developed agents are going to be able to be generalized in order to play well on different games.

Timeline of the competition:

June 1st, 2022: The first phase of warm-ups
June 10th, 2022: The second phase of warm-ups
June 20th, 2022: The first phase of the final rounds
June 30th, 2022: The second phase of the final rounds
July 15th, 2022: Announce winners

Competition url: http://www.jidiai.cn/ijcai_2022/

CC8: AIBIRDS 2022 – Angry Birds AI Competition

July 27 (Wednesday), 09:00-12:30 and 14:00-17:30, Schubert 1

The task of this competition is to develop a computer program that can successfully play Angry Birds. The long term goal is to build an intelligent Angry Birds playing agent that can play new levels better than the best human players. This is a difficult problem as it requires agents to predict the outcome of physical actions without having complete knowledge of the world, and then to select a good action out of infinitely many possible actions. This is an essential capability of future AI systems that interact with the physical world. The Angry Birds AI competition provides a simplified and controlled environment for developing and testing these capabilities. In addition to the Standard Track, in 2022 we again run the Novelty Track where we introduce novel components to the game during the competition (e.g., new game objects, or changes in properties of existing game objects) and agents are tasked with detecting when novelty occurs and to adjust their strategies to the novelty accordingly.

Timeline of the competition:

June 10th, 2022: Early registration
July 15th, 2022: Late Registration
July 2nd, 2022: Agent submission for testing (Novelty Track)
July 9th, 2022: : Agent submission for testing (Standard Track)
July 15th, 2022: Final Agent submission (Novelty Track)
July 22nd, 2022: Final Agent submission (Standard Track)

Competition url: http://aibirds.org/