Research Topic

Combining Symbolic Reasoning and Data-Driven Learning for Decision-Making

About this Research Topic

In recent times, deep network architectures and data-driven learning algorithms have come to represent state-of-the-art for many pattern recognition and control problems in robotics and AI. Examples of such problems include feature extraction, object recognition, visual question answering, robot navigation, ...

In recent times, deep network architectures and data-driven learning algorithms have come to represent state-of-the-art for many pattern recognition and control problems in robotics and AI. Examples of such problems include feature extraction, object recognition, visual question answering, robot navigation, and manipulation. However, these networks and algorithms require a large number of labeled training examples and substantial computational resources, which are difficult to obtain in many complex real-world domains. In addition, the internal representations and reasoning methods of the learned models are rather difficult to interpret, whereas this explainability is a key requirement for their use in many application domains.

An alternative strand of research in AI focuses on symbolic reasoning methods that represent and reason with incomplete prior information about the domain. This information includes commonsense knowledge, and information about domain dynamics and context, obtained from humans or prior experience of interacting with the environment. Many logical formalisms have been developed to represent this knowledge using hierarchical relational structures and rules, and to provide explainable reasoning with such representations. However, these methods often require considerable human input and domain expertise to encode domain knowledge. In addition, it is challenging to use these methods to represent and reason efficiently with noisy sensor input, and to provide accurate descriptions of complex, realistic environments.

With this Research Topic, we call for papers that exploit the complementary strengths of symbolic reasoning and data-driven techniques. The objective is to add to our understanding of the synergies between these methods and the associated branches of artificial intelligence, in the context of decision-making agents and robots.

Topics of interest include, but are not limited to:
• Cognitive systems
• Deep symbolic reinforcement learning
• Goal and intention recognition
• Human-robot/agent and multiagent/multirobot collaboration
• Interactive machine/task learning
• Knowledge representation and reasoning
• Planning and control algorithms
• Relational learning
• Symbol grounding and task representation
• Applications (computer vision, robotics, natural language processing, etc.)


Keywords: Hybrid AI, Symbolic Reasoning, Logic, Deep Learning


Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.

Recent Articles

Loading..

About Frontiers Research Topics

With their unique mixes of varied contributions from Original Research to Review Articles, Research Topics unify the most influential researchers, the latest key findings and historical advances in a hot research area! Find out more on how to host your own Frontiers Research Topic or contribute to one as an author.

Topic Editors

Loading..

Submission Deadlines

14 February 2020 Manuscript

Participating Journals

Manuscripts can be submitted to this Research Topic via the following journals:

Loading..

Topic Editors

Loading..

Submission Deadlines

14 February 2020 Manuscript

Participating Journals

Manuscripts can be submitted to this Research Topic via the following journals:

Loading..
Loading..

total views article views article downloads topic views

}
 
Top countries
Top referring sites
Loading..

Comments

Loading..

Add a comment

Add comment
Back to top