[robotics-worldwide] [meetings][CfP][Challenge] GOSEEK RL Challenge and ICRA 2020 workshop

Luca Carlone lcarlone at mit.edu
Sun Mar 8 17:57:57 PDT 2020


Dear all,

I’m happy to announce that the GOSEEK challenge is online! Please find instructions here: https://urldefense.com/v3/__https://github.com/MIT-TESSE/goseek-challenge__;!!LIr3w8kk_Xxm!7foDWt9g1RsSPdnqSN8R_74oLMeZ-VgZnRT4i7exU-mv8lGvRJxlVOZ5wP8VfxCdj1GSMQPj$ 

GOSEEK is a reinforcement learning challenge. The challenge consists in creating an RL agent that combines advanced perception (provided by Kimera<https://urldefense.com/v3/__https://arxiv.org/abs/1910.02490__;!!LIr3w8kk_Xxm!7foDWt9g1RsSPdnqSN8R_74oLMeZ-VgZnRT4i7exU-mv8lGvRJxlVOZ5wP8VfxCdj2gN5fQo$ >) and high-level decision-making to search for objects placed within complex indoor environments from a Unity-based simulator. Simply put: like PACMAN, but in a realistic scene and with realistic perception capabilities. Several data modalities are provided from both the simulator ground truth and the perception pipeline (e.g., images, depth, agent location) to enable the participants to focus on the RL/search aspects. The contest is hosted on the EvalAI platform, where participants will submit solutions, via docker containers run on AWS instances, for scoring. The winner of the competition will receive a monetary prize and will give a keynote presentation at the workshop.

To get updates about the challenge, please send an email to qla at mit.edu<mailto:qla at mit.edu>  with subject “GOSEEK: subscribe” and we will add you to our Goseek-Challenge at mit.edu<mailto:Goseek-Challenge at mit.edu>  mailing list.
The challenge is organized in conjunction with the ICRA 2020 workshop "Perception, Action, Learning: from Metric-Semantic Scene Understanding to High-level Task Execution".
Please find the call for paper and more details about the workshop below.

COVID-19 notice: We are closely monitoring any travel advisories and governmental recommendations, including the ones reported on the ICRA website (https://urldefense.com/v3/__https://www.icra2020.org/registration/coronavirus__;!!LIr3w8kk_Xxm!7foDWt9g1RsSPdnqSN8R_74oLMeZ-VgZnRT4i7exU-mv8lGvRJxlVOZ5wP8VfxCdjwZYxr9O$ ). While the outbreak should have minimal impact on the (virtual) competition, it might cause disruptions for the actual workshop.
(do you remember the good times in which a virus was more of a worry for a robot than for a human?)
We will provide updates on the workshop website and via the Goseek-Challenge at mit.edu<mailto:Goseek-Challenge at mit.edu> mailing list as things evolve.

Stay safe!

Luca

------------------------------------------------------------------------------------------------------------------------------------------------
ICRA 2020 Workshop and Challenge:
Perception, Action, Learning: from Metric-Semantic Scene Understanding to High-level Task Execution
International Conference on Robotics and Automation - May 31 to June 4 2020, Palais des Congrès de Paris - FRANCE
Website: https://urldefense.com/v3/__https://mit-spark.github.io/PAL-ICRA2020/__;!!LIr3w8kk_Xxm!7foDWt9g1RsSPdnqSN8R_74oLMeZ-VgZnRT4i7exU-mv8lGvRJxlVOZ5wP8VfxCdj9c_ymIN$ 
Submission Link:  https://urldefense.com/v3/__https://easychair.org/conferences/?conf=pal2020icraworkshop__;!!LIr3w8kk_Xxm!7foDWt9g1RsSPdnqSN8R_74oLMeZ-VgZnRT4i7exU-mv8lGvRJxlVOZ5wP8VfxCdj2CjAQt7$ 
Submission Deadline: March 30, 2020
Challenge Submission Deadline: May 1, 2020

OVERVIEW:
----------------
This workshop brings together researchers from robotics, computer vision, and machine learning to examine challenges and opportunities emerging at the boundary between spatial perception and high-level task execution. Recent years have seen a growing interest towards metric-semantic understanding, which consists in building a semantically annotated (or object-oriented) model of the environment. On the other hand, researchers have been looking at high-level task execution using modern tools from reinforcement learning and traditional decision-making. The combination of these research efforts in spatial perception and task execution has the potential to enable applications such as visual question-answering, object search and retrieval, and provides a more intuitive interaction with the user. This workshop creates an exchange opportunity to connect researchers working in metric-semantic perception and task execution. In particular, the workshop will bring forward the latest breakthroughs and cutting-edge research in the two research areas. Besides the usual mix of invited talks and poster presentations, the workshop involves two interactive activities. First, we will provide a hands-on tutorial on a state-of-the-art library for metric-semantic reconstruction (Kimera<https://urldefense.com/v3/__https://github.com/MIT-SPARK/Kimera__;!!LIr3w8kk_Xxm!7foDWt9g1RsSPdnqSN8R_74oLMeZ-VgZnRT4i7exU-mv8lGvRJxlVOZ5wP8VfxCdj_2GL2s5$ >). Second, we will organize the GOSEEK challenge (details below), in conjunction with the release of a photo-realistic Unity-based simulator, where participants will need to combine perception and decision-making to find objects in a complex indoor environment.
- The workshop will include keynote presentations from established researchers in robotics, machine learning, computer vision, robot perception.
- There will be two spotlight talks and two poster sessions highlighting contributed papers throughout the day.
- The winner of the challenge will give an invited keynote presentation  (best 3 take home a monetary prize).

The workshop is endorsed by the IEEE RAS Technical Committee for Computer & Robot Vision.

CHALLENGE AND AWARDS:
--------------
We are organizing the GOSEEK challenge, where participants create an RL agent that combines perception and high-level decision-making to search for objects placed within complex indoor environments from a Unity-based simulator. Simply put: like PACMAN, but in a realistic scene and with realistic perception capabilities. Several data modalities will be provided from both the simulator ground truth and a perception pipeline (e.g., images, depth, agent location) to enable the participants to focus on the RL/search aspects. The contest will be hosted on the EvalAI platform, where participants will submit solutions, via docker containers run on AWS instances, for scoring. The winner of the competition will receive a monetary prize and will give a keynote presentation at the workshop.
Tentative timeline:
- February 15: Challenge website is online with ground truth data.
- March 1: Full perception capabilities are made available. Challenge ready for submissions!
- May 1: Deadline to submit your RL agent!

SUBMISSIONS:
---------------------
Participants are invited to submit an extended abstract or short papers (up to 4 pages in ICRA format) focusing on novel advances in spatial perception, reinforcement learning, and at the boundary between these research areas.
Topics of interest include but are not limited to:
- Novel algorithms for spatial perception that combine geometry, semantics, and physics, and allow reasoning over spatial, semantic, and temporal aspects;
- Learning techniques that can produce cognitive representations directly from complex sensory inputs;
- Approaches that combine learning-based techniques with geometric and model-based estimation methods;
- Novel transfer learning and meta-learning methods for reinforcement learning;
- Novel RL approaches that leverage domain knowledge and existing (model-free and model-based) methods for perception and planning; and
- Position papers and unconventional ideas on how to reach human-level performance in robot perception and task-execution.
Contributed papers will be reviewed by the organizers and a program committee of invited reviewers. Accepted papers will be published on the workshop website and will be featured in spotlight presentations and poster sessions.

Submission link: https://urldefense.com/v3/__https://easychair.org/conferences/?conf=pal2020icraworkshop__;!!LIr3w8kk_Xxm!7foDWt9g1RsSPdnqSN8R_74oLMeZ-VgZnRT4i7exU-mv8lGvRJxlVOZ5wP8VfxCdj2CjAQt7$ 

IMPORTANT DATES:
----------------------------
- Submission Deadline: March 30, 2020
- Notification of Acceptance: April 30, 2020
- Workshop Date: May 31, 2020

INVITED SPEAKERS (tentative):
-----------------------------------
- Raia Hadsell (DeepMind)
- Dhruv Batra (Georgia Tech)
- Sertac Karaman (MIT)
- Andrew Davison (Imperial College)
- Cesar Cadena (ETH Zurich)
- Marco Pavone (Stanford)
- Davide Scaramuzza (University of Zurich)

ORGANIZING COMMITTEE:
-------------------------------------
- Luca Carlone, Massachusetts Institute of Technology
- Dan Griffith, Massachusetts Institute of Technology Lincoln Laboratory
- Sanjeev Mohindra, Massachusetts Institute of Technology Lincoln Laboratory

FURTHER INFORMATION:
-----------------------------------
Please send any questions to Luca Carlone (lcarlone at mit.edu<mailto:lcarlone at mit.edu>). Please include "PAL ICRA 2020 Workshop" in the subject of the email.

------------------------------------------------------------------------------------
Luca Carlone
Charles Stark Draper Assistant Professor
Laboratory for Information and Decision Systems (LIDS)
Massachusetts Institute of Technology (MIT)
office: 32 Vassar St., Cambridge, MA 02139, Room: 31-243
web: https://urldefense.com/v3/__http://www.lucacarlone.com/__;!!LIr3w8kk_Xxm!7foDWt9g1RsSPdnqSN8R_74oLMeZ-VgZnRT4i7exU-mv8lGvRJxlVOZ5wP8VfxCdj5pcsAF6$ 
------------------------------------------------------------------------------------




More information about the robotics-worldwide mailing list