AWICS: Advancing Women in Computer Science

This effort has a management committee that includes two key TRIPODS players: Fred Roberts and Matthew Stone. Visit https://awics.cs.rutgers.edu/.

AI & Pandemics

This effort has an organizing committee that includes three key TRIPODS players, Kostas Bekris and David Pennock and Fred Roberts. Other TRIPODS faculty are also involved: Konstantin Mischaikow and Ying Hung and Matthew Stone. Visit https://pandemic.dimacs.rutgers.edu/.

Data Science (DS)

Data Science (DS) is an interdisciplinary program of study, housed in the Rutgers-New Brunswick (RU-NB) School of Arts and Sciences (SAS): https://mps.rutgers.edu/data-science. It is offered jointly by the Departments of Computer Science (CS) and Department of Statistics (Stat) in partnership with the RU-NB School of Communication and Information (SC&I) and other SAS departments. It is designed to equip students to understand the basic principles of computation, statistical inference, and data management, and their applications in a specific domain/field. It is open to ALL students in any RU-NB program.

Machine Learning challenge for Combinatorial Problems in Robotics

We invite you to participate in a Machine Learning challenge for Combinatorial Problems in Robotics, which has been setup on Kaggle:

https://www.kaggle.com/c/object-rearrangement-competition

This challenge provides the opportunity to develop data-driven methods for making predictions to help solve intriguing object rearrangement problems. Object rearrangement is a fundamental problem in many everyday tasks, e.g., tidying up a desk, organizing a shelf, or sorting products. For the same reason, the problem is of key interest in the development of robotic manipulation solutions. The problem considered by the competition involves a set of objects within a bounded 2D workspace, which must be moved from one configuration to another. Only one object can be moved at a time to mimic the setup where the objects are being picked and placed by a robotic arm. Objects are not permitted to collide with each other. The typical objective is to realize a rearrangement with the least number of object moves.

Given prior work, we have decomposed this problem into three classification subtasks, which can be approached with machine learning tools. We have provided training and test data for these classification challenges via Kaggle. Please refer to the attached PDF document to read the description of the three challenges, how to access the data and how to submit your solutions.

As this is the initial release of the dataset and the competition, we will highly appreciate any feedback on how to improve the available information and simplify the process for accessing the data and submitting solutions.

What is in it for you?
You will gain experience in solving an interesting problem at the intersection of Combinatorics, Machine Learning and Robotics.
If you significantly contribute to the improvement of the competition’s information and management with feedback: we would like to invite you to join in research activities of the CS Robotics groups.
If you can come up with a machine learning solution for successfully tackling the challenge: we would like to invite you to join us in writing a research paper to be submitted to a flagship robotics conference, which could benefit your job search. We will also provide recommendation letters for your job search.

Best,
Kostas Bekris – on behalf of the ORLC co-organizers: Dan Nakhimovich, Kai Gao, Rui Wang, Jingjin Yu, Fred Roberts

PS: This dataset is being developed with the support of DATA-INSPIRE, an NSF-supported TRIPODS institute at Rutgers under award #1934924. More Info