Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Dorsa Sadigh

Dorsa Sadigh

Emergent Prosociality in Multi-Agent Games Through Gifting


May 13, 2021
Woodrow Z. Wang, Mark Beliaev, Erdem Bıyık, Daniel A. Lazar, Ramtin Pedarsani, Dorsa Sadigh

Figure 1 for Emergent Prosociality in Multi-Agent Games Through Gifting
Figure 2 for Emergent Prosociality in Multi-Agent Games Through Gifting
Figure 3 for Emergent Prosociality in Multi-Agent Games Through Gifting
Figure 4 for Emergent Prosociality in Multi-Agent Games Through Gifting
* 9 pages, 6 figures, IJCAI 2021 
  Access Paper or Ask Questions

Learning Visually Guided Latent Actions for Assistive Teleoperation


May 02, 2021
Siddharth Karamcheti, Albert J. Zhai, Dylan P. Losey, Dorsa Sadigh

Figure 1 for Learning Visually Guided Latent Actions for Assistive Teleoperation
Figure 2 for Learning Visually Guided Latent Actions for Assistive Teleoperation
Figure 3 for Learning Visually Guided Latent Actions for Assistive Teleoperation
Figure 4 for Learning Visually Guided Latent Actions for Assistive Teleoperation
* Accepted at Learning for Dynamics and Control (L4DC) 2021. 12 pages, 4 figures 
  Access Paper or Ask Questions

On the Critical Role of Conventions in Adaptive Human-AI Collaboration


Apr 07, 2021
Andy Shih, Arjun Sawhney, Jovana Kondic, Stefano Ermon, Dorsa Sadigh

Figure 1 for On the Critical Role of Conventions in Adaptive Human-AI Collaboration
Figure 2 for On the Critical Role of Conventions in Adaptive Human-AI Collaboration
Figure 3 for On the Critical Role of Conventions in Adaptive Human-AI Collaboration
Figure 4 for On the Critical Role of Conventions in Adaptive Human-AI Collaboration
* 9th International Conference on Learning Representations (ICLR 2021) 
  Access Paper or Ask Questions

Learning Human Objectives from Sequences of Physical Corrections


Mar 31, 2021
Mengxi Li, Alper Canberk, Dylan P. Losey, Dorsa Sadigh

Figure 1 for Learning Human Objectives from Sequences of Physical Corrections
Figure 2 for Learning Human Objectives from Sequences of Physical Corrections
Figure 3 for Learning Human Objectives from Sequences of Physical Corrections
Figure 4 for Learning Human Objectives from Sequences of Physical Corrections
* 7 pages, 4 figures 
  Access Paper or Ask Questions

Learning from Imperfect Demonstrations from Agents with Varying Dynamics


Mar 10, 2021
Zhangjie Cao, Dorsa Sadigh

Figure 1 for Learning from Imperfect Demonstrations from Agents with Varying Dynamics
Figure 2 for Learning from Imperfect Demonstrations from Agents with Varying Dynamics
Figure 3 for Learning from Imperfect Demonstrations from Agents with Varying Dynamics
Figure 4 for Learning from Imperfect Demonstrations from Agents with Varying Dynamics
* Accpeted by ICRA 2021 
  Access Paper or Ask Questions

ELLA: Exploration through Learned Language Abstraction


Mar 10, 2021
Suvir Mirchandani, Siddharth Karamcheti, Dorsa Sadigh

Figure 1 for ELLA: Exploration through Learned Language Abstraction
Figure 2 for ELLA: Exploration through Learned Language Abstraction
Figure 3 for ELLA: Exploration through Learned Language Abstraction
Figure 4 for ELLA: Exploration through Learned Language Abstraction
* 13 pages, 8 figures 
  Access Paper or Ask Questions

Transfer Reinforcement Learning across Homotopy Classes


Feb 11, 2021
Zhangjie Cao, Minae Kwon, Dorsa Sadigh

Figure 1 for Transfer Reinforcement Learning across Homotopy Classes
Figure 2 for Transfer Reinforcement Learning across Homotopy Classes
Figure 3 for Transfer Reinforcement Learning across Homotopy Classes
Figure 4 for Transfer Reinforcement Learning across Homotopy Classes
* Accepted by IEEE Robotics and Automation Letters 2021 
  Access Paper or Ask Questions

Incentivizing Routing Choices for Safe and Efficient Transportation in the Face of the COVID-19 Pandemic


Dec 28, 2020
Mark Beliaev, Erdem Bıyık, Daniel A. Lazar, Woodrow Z. Wang, Dorsa Sadigh, Ramtin Pedarsani

Figure 1 for Incentivizing Routing Choices for Safe and Efficient Transportation in the Face of the COVID-19 Pandemic
Figure 2 for Incentivizing Routing Choices for Safe and Efficient Transportation in the Face of the COVID-19 Pandemic
Figure 3 for Incentivizing Routing Choices for Safe and Efficient Transportation in the Face of the COVID-19 Pandemic
* To appear at ICCPS 2021. 11 pages, 4 figures 
  Access Paper or Ask Questions

Learning Latent Representations to Influence Multi-Agent Interaction


Nov 12, 2020
Annie Xie, Dylan P. Losey, Ryan Tolsma, Chelsea Finn, Dorsa Sadigh

Figure 1 for Learning Latent Representations to Influence Multi-Agent Interaction
Figure 2 for Learning Latent Representations to Influence Multi-Agent Interaction
Figure 3 for Learning Latent Representations to Influence Multi-Agent Interaction
Figure 4 for Learning Latent Representations to Influence Multi-Agent Interaction
* Conference on Robot Learning (CoRL) 2020. Supplementary website at https://sites.google.com/view/latent-strategies/ 
  Access Paper or Ask Questions

ROIAL: Region of Interest Active Learning for Characterizing Exoskeleton Gait Preference Landscapes


Nov 09, 2020
Kejun Li, Maegan Tucker, Erdem Bıyık, Ellen Novoseller, Joel W. Burdick, Yanan Sui, Dorsa Sadigh, Yisong Yue, Aaron D. Ames

Figure 1 for ROIAL: Region of Interest Active Learning for Characterizing Exoskeleton Gait Preference Landscapes
Figure 2 for ROIAL: Region of Interest Active Learning for Characterizing Exoskeleton Gait Preference Landscapes
Figure 3 for ROIAL: Region of Interest Active Learning for Characterizing Exoskeleton Gait Preference Landscapes
Figure 4 for ROIAL: Region of Interest Active Learning for Characterizing Exoskeleton Gait Preference Landscapes
* 6 pages; 7 figures 
  Access Paper or Ask Questions

Learning Adaptive Language Interfaces through Decomposition


Oct 11, 2020
Siddharth Karamcheti, Dorsa Sadigh, Percy Liang

Figure 1 for Learning Adaptive Language Interfaces through Decomposition
Figure 2 for Learning Adaptive Language Interfaces through Decomposition
Figure 3 for Learning Adaptive Language Interfaces through Decomposition
Figure 4 for Learning Adaptive Language Interfaces through Decomposition
* Accepted at the 1st Workshop for Interactive and Executable Semantic Parsing (IntEx-SemPar) @ EMNLP 2020. 11 pages, 5 figures 
  Access Paper or Ask Questions

Multi-Agent Safe Planning with Gaussian Processes


Aug 10, 2020
Zheqing Zhu, Erdem Bıyık, Dorsa Sadigh

Figure 1 for Multi-Agent Safe Planning with Gaussian Processes
Figure 2 for Multi-Agent Safe Planning with Gaussian Processes
Figure 3 for Multi-Agent Safe Planning with Gaussian Processes
* 9 pages, 5 figures. Published at IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2020 
  Access Paper or Ask Questions

Learning User-Preferred Mappings for Intuitive Robot Control


Jul 22, 2020
Mengxi Li, Dylan P. Losey, Jeannette Bohg, Dorsa Sadigh

Figure 1 for Learning User-Preferred Mappings for Intuitive Robot Control
Figure 2 for Learning User-Preferred Mappings for Intuitive Robot Control
Figure 3 for Learning User-Preferred Mappings for Intuitive Robot Control
Figure 4 for Learning User-Preferred Mappings for Intuitive Robot Control
* 8 pages, 7 figures, Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), October 2020 
  Access Paper or Ask Questions

Reinforcement Learning based Control of Imitative Policies for Near-Accident Driving


Jul 01, 2020
Zhangjie Cao, Erdem Bıyık, Woodrow Z. Wang, Allan Raventos, Adrien Gaidon, Guy Rosman, Dorsa Sadigh

Figure 1 for Reinforcement Learning based Control of Imitative Policies for Near-Accident Driving
Figure 2 for Reinforcement Learning based Control of Imitative Policies for Near-Accident Driving
Figure 3 for Reinforcement Learning based Control of Imitative Policies for Near-Accident Driving
Figure 4 for Reinforcement Learning based Control of Imitative Policies for Near-Accident Driving
* 10 pages, 7 figures. Published at Robotics: Science and Systems (RSS) 2020 
  Access Paper or Ask Questions

Formalizing and Guaranteeing* Human-Robot Interaction


Jun 30, 2020
Hadas Kress-Gazit, Kerstin Eder, Guy Hoffman, Henny Admoni, Brenna Argall, Ruediger Ehlers, Christoffer Heckman, Nils Jansen, Ross Knepper, Jan Křetínský, Shelly Levy-Tzedek, Jamy Li, Todd Murphey, Laurel Riek, Dorsa Sadigh

Figure 1 for Formalizing and Guaranteeing* Human-Robot Interaction

  Access Paper or Ask Questions

Learning Reward Functions from Diverse Sources of Human Feedback: Optimally Integrating Demonstrations and Preferences


Jun 24, 2020
Erdem Bıyık, Dylan P. Losey, Malayandi Palan, Nicholas C. Landolfi, Gleb Shevchuk, Dorsa Sadigh

Figure 1 for Learning Reward Functions from Diverse Sources of Human Feedback: Optimally Integrating Demonstrations and Preferences
Figure 2 for Learning Reward Functions from Diverse Sources of Human Feedback: Optimally Integrating Demonstrations and Preferences
Figure 3 for Learning Reward Functions from Diverse Sources of Human Feedback: Optimally Integrating Demonstrations and Preferences
Figure 4 for Learning Reward Functions from Diverse Sources of Human Feedback: Optimally Integrating Demonstrations and Preferences
* 19 pages, 17 figures. Submitted to The International Journal of Robotics Research (IJRR) 
  Access Paper or Ask Questions

Active Preference-Based Gaussian Process Regression for Reward Learning


Jun 03, 2020
Erdem Bıyık, Nicolas Huynh, Mykel J. Kochenderfer, Dorsa Sadigh

Figure 1 for Active Preference-Based Gaussian Process Regression for Reward Learning
Figure 2 for Active Preference-Based Gaussian Process Regression for Reward Learning
Figure 3 for Active Preference-Based Gaussian Process Regression for Reward Learning
Figure 4 for Active Preference-Based Gaussian Process Regression for Reward Learning
* Proceedings of Robotics: Science and Systems (RSS), July 2020 
  Access Paper or Ask Questions

Dynamic Multi-Robot Task Allocation under Uncertainty and Temporal Constraints


May 29, 2020
Shushman Choudhury, Jayesh K. Gupta, Mykel J. Kochenderfer, Dorsa Sadigh, Jeannette Bohg

Figure 1 for Dynamic Multi-Robot Task Allocation under Uncertainty and Temporal Constraints
Figure 2 for Dynamic Multi-Robot Task Allocation under Uncertainty and Temporal Constraints
Figure 3 for Dynamic Multi-Robot Task Allocation under Uncertainty and Temporal Constraints
Figure 4 for Dynamic Multi-Robot Task Allocation under Uncertainty and Temporal Constraints
* Robotics Science and Systems (RSS) 2020; Source code at https://github.com/sisl/SCoBA.jl 
  Access Paper or Ask Questions

Shared Autonomy with Learned Latent Actions


May 11, 2020
Hong Jun Jeon, Dylan P. Losey, Dorsa Sadigh

Figure 1 for Shared Autonomy with Learned Latent Actions
Figure 2 for Shared Autonomy with Learned Latent Actions
Figure 3 for Shared Autonomy with Learned Latent Actions
Figure 4 for Shared Autonomy with Learned Latent Actions

  Access Paper or Ask Questions

BLEU Neighbors: A Reference-less Approach to Automatic Evaluation


Apr 29, 2020
Kawin Ethayarajh, Dorsa Sadigh

Figure 1 for BLEU Neighbors: A Reference-less Approach to Automatic Evaluation
Figure 2 for BLEU Neighbors: A Reference-less Approach to Automatic Evaluation
Figure 3 for BLEU Neighbors: A Reference-less Approach to Automatic Evaluation
Figure 4 for BLEU Neighbors: A Reference-less Approach to Automatic Evaluation

  Access Paper or Ask Questions

Exchangeable Input Representations for Reinforcement Learning


Mar 19, 2020
John Mern, Dorsa Sadigh, Mykel J. Kochenderfer

Figure 1 for Exchangeable Input Representations for Reinforcement Learning
Figure 2 for Exchangeable Input Representations for Reinforcement Learning
Figure 3 for Exchangeable Input Representations for Reinforcement Learning
Figure 4 for Exchangeable Input Representations for Reinforcement Learning
* 6 pages, 7 figures 
  Access Paper or Ask Questions

When Humans Aren't Optimal: Robots that Collaborate with Risk-Aware Humans


Jan 13, 2020
Minae Kwon, Erdem Biyik, Aditi Talati, Karan Bhasin, Dylan P. Losey, Dorsa Sadigh

Figure 1 for When Humans Aren't Optimal: Robots that Collaborate with Risk-Aware Humans
Figure 2 for When Humans Aren't Optimal: Robots that Collaborate with Risk-Aware Humans
Figure 3 for When Humans Aren't Optimal: Robots that Collaborate with Risk-Aware Humans
Figure 4 for When Humans Aren't Optimal: Robots that Collaborate with Risk-Aware Humans
* ACM/IEEE International Conference on Human-Robot Interaction 
  Access Paper or Ask Questions

Continual adaptation for efficient machine communication


Nov 22, 2019
Robert D. Hawkins, Minae Kwon, Dorsa Sadigh, Noah D. Goodman

Figure 1 for Continual adaptation for efficient machine communication
Figure 2 for Continual adaptation for efficient machine communication

  Access Paper or Ask Questions

Learning from My Partner's Actions: Roles in Decentralized Robot Teams


Oct 28, 2019
Dylan P. Losey, Mengxi Li, Jeannette Bohg, Dorsa Sadigh

Figure 1 for Learning from My Partner's Actions: Roles in Decentralized Robot Teams
Figure 2 for Learning from My Partner's Actions: Roles in Decentralized Robot Teams
Figure 3 for Learning from My Partner's Actions: Roles in Decentralized Robot Teams
Figure 4 for Learning from My Partner's Actions: Roles in Decentralized Robot Teams
* Conference on Robot Learning (CoRL), 2019 
  Access Paper or Ask Questions

Controlling Assistive Robots with Learned Latent Actions


Oct 16, 2019
Dylan P. Losey, Krishnan Srinivasan, Ajay Mandlekar, Animesh Garg, Dorsa Sadigh

Figure 1 for Controlling Assistive Robots with Learned Latent Actions
Figure 2 for Controlling Assistive Robots with Learned Latent Actions
Figure 3 for Controlling Assistive Robots with Learned Latent Actions
Figure 4 for Controlling Assistive Robots with Learned Latent Actions

  Access Paper or Ask Questions

Asking Easy Questions: A User-Friendly Approach to Active Reward Learning


Oct 10, 2019
Erdem Bıyık, Malayandi Palan, Nicholas C. Landolfi, Dylan P. Losey, Dorsa Sadigh

Figure 1 for Asking Easy Questions: A User-Friendly Approach to Active Reward Learning
Figure 2 for Asking Easy Questions: A User-Friendly Approach to Active Reward Learning
Figure 3 for Asking Easy Questions: A User-Friendly Approach to Active Reward Learning
Figure 4 for Asking Easy Questions: A User-Friendly Approach to Active Reward Learning
* Proceedings of the 3rd Conference on Robot Learning (CoRL), October 2019 
  Access Paper or Ask Questions