Contents
18 December 2019
Vol 4, Issue 37
Vol 4, Issue 37
Focus
- XAI—Explainable artificial intelligence
Explainability is essential for users to effectively understand, trust, and manage powerful artificial intelligence applications.
- Explainable robotics in science fiction
Would we trust a robot? Science fiction says no, but explainable robotics may find a way.
Research Articles
- A tale of two explanations: Enhancing human trust by explaining robot behavior
Forms of explanation that are best suited to foster trust do not necessarily correspond to those components contributing to the best task performance.
- A formal methods approach to interpretable reinforcement learning for robotic planning
A formal methods approach to reinforcement learning generates rewards from a formal language and guarantees safety.
- An autonomous untethered fast soft robotic insect driven by low-voltage dielectric elastomer actuators
Low-voltage dielectric elastomer actuators propel an autonomous untethered subgram robust soft legged robot along complex paths.
About The Cover
ONLINE COVER Explainable Artificial Intelligence (XAI). Recent successes in machine learning have lead to new AI applications that offer important benefits. However, many of these systems are not able to explain their autonomous decisions and actions to humans, a critical ability in areas such as defense, medicine, and finance. Edmonds et al. used a robotic system that could learn to open bottles with various safety mechanisms. The robot then demonstrated this ability to people with various explanations, and the authors noted which explanations best fostered trust. Li et al. developed an interpretable, safe approach to reinforcement learning that was then applied to robots cooking and serving hot dogs. [CREDIT: IMAGINIMA/ISTOCK.COM]
