Back to Search
Start Over
Combining deep reinforcement learning with prior knowledge and reasoning
- Source :
- ACM SIGAPP Applied Computing Review. 18:33-45
- Publication Year :
- 2018
- Publisher :
- Association for Computing Machinery (ACM), 2018.
-
Abstract
- Recent improvements in deep reinforcement learning have allowed to solve problems in many 2D domains such as Atari games. However, in complex 3D environments, numerous learning episodes are required which may be too time consuming or even impossible especially in real-world scenarios. We present a new architecture to combine external knowledge and deep reinforcement learning using only visual input. A key concept of our system is augmenting image input by adding environment feature information and combining two sources of decision. We evaluate the performances of our method in 3D partially-observable environments from the Microsoft Malmo platform. Experimental evaluation exhibits higher performance and faster learning compared to a single reinforcement learning model.
- Subjects :
- Computer science
business.industry
Deep learning
Cognitive neuroscience of visual object recognition
Ocean Engineering
02 engineering and technology
Machine learning
computer.software_genre
Image (mathematics)
03 medical and health sciences
0302 clinical medicine
Feature (computer vision)
0202 electrical engineering, electronic engineering, information engineering
Key (cryptography)
Reinforcement learning
020201 artificial intelligence & image processing
Artificial intelligence
Architecture
business
computer
030217 neurology & neurosurgery
Subjects
Details
- ISSN :
- 19310161 and 15596915
- Volume :
- 18
- Database :
- OpenAIRE
- Journal :
- ACM SIGAPP Applied Computing Review
- Accession number :
- edsair.doi...........0d11e2aae7f05445382a4a4f5160d7e0