Cost Function

Cost functions are mathematical representations that quantify the desirability of different outcomes in optimization problems, crucial for guiding intelligent systems towards optimal behavior. Current research focuses on learning these functions from data, often employing neural networks (like graph neural networks and Transformers) within frameworks such as inverse reinforcement learning and model predictive control. This allows for the adaptation of cost functions to complex, real-world scenarios, such as robotic manipulation, autonomous driving, and human-robot interaction, improving the performance and safety of these systems. The ability to learn effective cost functions directly impacts the development of robust and adaptable AI agents across diverse applications.

Papers

July 17, 2024

Robotic Arm Manipulation with Inverse Reinforcement Learning & TD-MPC
Md Shoyib Hassan, Sabir Md Sanaullah
Inverse Reinforcement Learning Manipulation Task Td MPC Cost Function

July 15, 2024

Learning Social Cost Functions for Human-Aware Path Planning
Andrea Eirale, Matteo Leonetti, Marcello Chiaberge
Social Navigation Robotic Agent Social Robot Navigation Cost Function Social Scenario

November 14, 2023

SceneScore: Learning a Cost Function for Object Arrangement
Ivan Kapelyukh, Edward Johns
Non Humanoid Robot Value Function Object Relation Cost Function Object Arrangement

October 9, 2023

DTPP: Differentiable Joint Conditional Prediction and Cost Evaluation for Tree Policy Planning in Autonomous Driving
Zhiyu Huang, Peter Karkus, Boris Ivanovic, Yuxiao Chen, Marco Pavone, Chen Lv
Autonomous Driving Ego Motion Joint predictioN Joint Distribution Adaptation Cost Function

May 4, 2023

Maximum Causal Entropy Inverse Constrained Reinforcement Learning
Mattijs Baert, Pietro Mazzaglia, Sam Leroux, Pieter Simoens
Inverse Reinforcement Learning Artificial Agent State Action Space Learning Agent Causal Entropy Cost Function

February 23, 2023

A Statistical Learning Take on the Concordance Index for Survival Analysis
Alex Nowak-Vila, Kevin Elgui, Genevieve Robin
Machine Learning Statistical Learning Survival Model Cost Function Concordance Index

January 31, 2023

Learning Risk-Aware Costmaps via Inverse Reinforcement Learning for Off-Road Navigation
Samuel Triest, Mateo Guaman Castro, Parv Maheshwari, Matthew Sivaprakasam, Wenshan Wang, Sebastian Scherer
Inverse Reinforcement Learning Local Navigation Road Navigation Road Driving Cost Function Cost Map

January 24, 2023

AutoCost: Evolving Intrinsic Cost for Zero-violation Reinforcement Learning
Tairan He, Weiye Zhao, Changliu Liu
Deep Reinforcement Learning Hidden CoST Constrained Reinforcement Learning Constrained Markov Decision Process Cost Function Constrained Reinforcement LeArning

July 21, 2022

Differentiable Integrated Motion Prediction and Planning with Learnable Cost Function for Autonomous Driving
Zhiyu Huang, Haochen Liu, Jingda Wu, Chen Lv
Autonomous Driving Autonomous Vehicle Task Planning Urban Driving Autonomous Driving System Prediction Boosted Planning Framework Cost Function

March 30, 2022

Learning of Global Objective for Network Flow in Multi-Object Tracking
Shuai Li, Yu Kong, Hamid Rezatofighi
LeArning Abstract Multi Object Tracking Tracking Benchmark Network Flow Cost Function Cost Flow