We invite proposals for the International Conference on Automated Planning and Scheduling 2021 (ICAPS 2021) competition track to be held in Guangzhou, China. We solicit competition proposals on any topic of interest to the Planning & Scheduling community. We especially encourage competition proposals from emerging fields or new application domains related to Planning and Scheduling. Interdisciplinary topics that will attract a significant cross-section of the community are especially encouraged.
Participants will compete to obtain the best score on a planning task of interest to the ICAPS community based on the problem and data defined and released by the organizers of the competition. Organizers for a certain challenge should define the tasks specifically such that participants can build their planning/scheduling models and/or their learning models based on the task descriptions. Example challenge problems for competitions are (but certainly not limited to):
- Bicycle sharing: A bicycle-sharing system is a service in which bicycles are made available for shared use to individuals on a short term basis. Many bike share systems allow people to borrow a bike from a "dock" and return it at another dock belonging to the same system. To reduce the loss of user demands on using bicycles in every station, the system needs to make a plan to reschedule bicycles among stations by predicting future use of bicycles in each station. Considering there are many constraints related to actions of repositioning bicycles, solving tasks of repositioning bicycles with (probabilistic) planning or scheduling models is likely to improve the effectiveness and efficiency.
- Smart dispatching: A smart-dispatching system is to effectively and efficiently dispatch a service, e.g., a driver in DIDI or Uber, to a passenger’s demand. The system needs to consider the current locations of drivers and passengers, the destinations of passengers, and the current status of traffic. In this scenario, planning and scheduling techniques could be adopted to improve the efficiency of dispatching considering the complexity of factors and the large scale of demands and supplies, and exhibit explanations when confronting unexpected issues, such as some clients waiting too long or lots of complaints received.
- Auto-driving Auto-driving needs to handle complicated uncertain environments rapidly. Large-scale observations from sensors often makes auto-driving systems difficult to make accurate decisions efficiently. Uncertain/Probabilistic planning and path planning could be helpful for improving auto-driving systems.
- Unmanned warehouse: Sorting packages in an unmanned warehouse is a challenging planning task. It needs to consider various packages in the same order and make efficient routes for each order such that each order can be delivered efficiently. It is possible to model the package sorting problem as a planning or scheduling problem and explore planning techniques to improve the effectiveness and efficiency.
- Chemical synthesis: Synthesizing small organic molecules is a process of planning a series of reactions. It is challenging to synthesize organic molecules of high-quality with respect to some specific features, considering the large scale of reactions and molecules. Automated retrosynthesis or synthesis of molecules has been an open problem for 50 years due to its scientific importance in chemistry as well as its direct applications to pharmaceutical and material industries.
There will be a competition track at ICAPS 2021, co-located with the conference workshops, where competition results will be presented and discussed. Competition organizers will be asked to propose a tentative schedule for presentation of the competition and its results during an assigned time slot; ICAPS will provide coffee breaks and, if necessary, poster facilities.
For any additional questions please contact the ICAPS competition chairs.
PROPOSAL SUBMISSIONThe proposal should include at least five parts:
- 1. Detailed task description;
- 2. Comprehensive competition rules;
- 3. Description of environment or simulator, data, code, and so on, and their readiness;
- 4. Deadlines;
- 5. Awards.
Proposers are also welcome to submit details of the task as potential challenge problems to the ICAPS Applications Track. Please consult the corresponding CFP for more details.
REVIEWING AND SELECTION PROCESSA competition program committee will be formed, consisting of experts on planning, learning and challenge organization. Each proposal will be reviewed by at least three members of the committee. The factors that will be considered when evaluating proposals include:
- Task: Impact, originality, relevance to the ICAPS community will all be considered. Tasks that include humanitarian and/or positive societal impact are especially highly encouraged, although other topics relevant to the ICAPS community will also be considered. If humanitarian projects are submitted, the involvement of the community in question is strongly desired, and “parachute science” is strongly discouraged.
- Protocol and Readiness: Feasibility of the task chosen, soundness of the evaluation criteria, and clarity and fairness of the competition rules will be considered. The readiness of the competition in terms of data, code, etc. will also be considered.
- Logistics: Schedule, plan for attracting competition participants, and experience and diversity of the organizers will all be considered. The specific plan for attracting participants of groups under-represented at ICAPS will be considered.
- Competition proposal submission deadline: Jan 26, 2021
- Acceptance notification: Jan 31, 2021
- Competition track: Jun 7-12, 2021 The reference timezone for all deadlines is UTC-12.
- Competition organizers should propose a timeline for running the competition to ensure that participants have enough time to contribute high-quality entries. It is recommended that competitions be completed by end of May, 2021 at the absolute latest.
- Competition organizers that require help or suggestions regarding competition platforms for running the competition can contact the competition chairs for advice.
COMPETITION CHAIRSTathagata Charkraboti (IBM Research AI)
Rong Pan (Sun Yat-Sen University)
Jianye Hao (Tianjin University)