Intermediate Target

Intermediate targets are proving crucial in improving the performance of various machine learning tasks, particularly in complex sequential decision-making problems. Current research focuses on integrating intermediate targets into reinforcement learning and other frameworks, such as using waypoints in trajectory planning for autonomous vehicles or discrete units in spoken language understanding, to guide learning and enhance model robustness. This approach addresses limitations of end-to-end methods by providing structured guidance and improving the ability to handle noisy or incomplete data, leading to more efficient and effective algorithms across diverse applications. The resulting improvements in performance and efficiency have significant implications for fields ranging from robotics and autonomous systems to natural language processing.

Papers

June 12, 2024

Utilizing Navigation Paths to Generate Target Points for Enhanced End-to-End Autonomous Driving Planning
Yuanhua Shen, Jun Li
Autonomous Vehicle End to End Driving Planning Top Quality Planning Intermediate Target

May 8, 2024

Evolving R2 to R2+: Optimal, Delayed Line-of-sight Vector-based Path Planning
Yan Kai Lai, Prahlad Vadakkepat, Cheng Xiang
Path Planning Line of Sight Obstacle Scenario Angle Path Intermediate Target

November 29, 2023

Is Inverse Reinforcement Learning Harder than Standard Reinforcement Learning? A Theoretical Perspective
Lei Zhao, Mengdi Wang, Yu Bai
Reward Function Inverse Reinforcement Learning Standard Reinforcement Learning Intermediate Target

June 24, 2023

Waypoint Transformer: Reinforcement Learning via Supervised Learning with Intermediate Targets
Anirudhan Badrinath, Yannis Flet-Berliac, Allen Nie, Emma Brunskill
Reinforcement Learning Offline Reinforcement Learning Supervised Learning Decision Transformer Waypoint Generation Intermediate Target

May 29, 2023

Improving Textless Spoken Language Understanding with Discrete Units as Intermediate Target
Guan-Wei Wu, Guan-Ting Lin, Shang-Wen Li, Hung-yi Lee
Spoken Language Understanding Self Supervised Speech Model Pre Trained Automatic Speech Recognition End to End Spoken Language Speech Text Data Intermediate Target

Intermediate Target

Papers

Utilizing Navigation Paths to Generate Target Points for Enhanced End-to-End Autonomous Driving Planning

Evolving R2 to R2+: Optimal, Delayed Line-of-sight Vector-based Path Planning

Is Inverse Reinforcement Learning Harder than Standard Reinforcement Learning? A Theoretical Perspective

Waypoint Transformer: Reinforcement Learning via Supervised Learning with Intermediate Targets

Improving Textless Spoken Language Understanding with Discrete Units as Intermediate Target