About this Research Topic
The rise in the adoption of artificial intelligence techniques is highlighting their shortcomings in terms of robustness, interpretability, usability, and trustworthiness. Challenges that need to be addressed to make advances in the next generation of AI systems arise from both computation and interaction standpoints. From the computational perspective, AI systems are dominated by machine-learned models that often inherit biases from the training data and generate undesired outcomes with safety, ethical, and societal concerns across applications. From the interactional perspective, the inconsistent and sometimes discriminative behavior of AI systems violates established usability guidelines and hurts user trust. A promising approach to address both computational and interactional challenges while building AI systems is the use of human-in-the-loop approaches (e.g crowd computing), which offer viable means to engage a large number of human participants in data-related tasks and in user studies.
Existing research and practice have mainly focused on leveraging crowd computing for training data creation. The perspective is rather limiting in terms of how AI can fully benefit from human-in-the-loop approaches. In the context of overcoming the computational and interactional challenges facing the current generation of AI systems, recent work has shown how crowd computing can be leveraged to either debug noisy training data in machine learning systems, understand which machine learning models are more congruent to human understanding in particular tasks, or to advance our understanding of how AI systems can influence human behavior. The goal of this call is to elicit new research outcomes that address important computational and interactional challenges to make advances in the next generation of AI systems.
Possible submission topics include:
- Human-in-the-loop AI
- Human-AI interaction
- Crowdsourced data and knowledge creation for AI
- Socio-technical aspects of AI systems: privacy, bias, trust
- Worker and user protection
- Conversational interfaces for human-AI collaboration
- Accountability and transparency in AI algorithms
- Ensuring fairness in AI
- AI promoted societal equity
- Operationalisation of ethical principles
Keywords: Human-in-the-Loop AI, Human-AI Interaction, Human Computation and Crowdsourcing, Responsible AI, Reliable AI, Interpretable AI, Ethics and AI, AI for Social Good
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.