TY  - JOUR
T1  - Hybrid Learning for Vision-and-Language Navigation Agents
AU  - Oh, Suntaek 
AU  - Kim, Incheol 
JO  - KIPS Transactions on Software and Data Engineering
PY  - 2020
DA  - 2020/1/30
DO  - https://doi.org/10.3745/KTSDE.2020.9.9.281
KW  - Vision-and-Language Navigation
KW  - Hybrid Learning
KW  - Path-Based Reward Function
AB  - The Vision-and-Language Navigation(VLN) task is a complex intelligence problem that requires both visual and language comprehension skills. In this paper, we propose a new learning model for visual-language navigation agents. The model adopts a hybrid learning that combines imitation learning based on demo data and reinforcement learning based on action reward. Therefore, this model can meet both problems of imitation learning that can be biased to the demo data and reinforcement learning with relatively low data efficiency. In addition, the proposed model uses a novel path-based reward function designed to solve the problem of existing goal-based reward functions. In this paper, we demonstrate the high performance of the proposed model through various experiments using both Matterport3D simulation environment and R2R benchmark dataset.