Intelligible Artificial Intelligence
نویسندگان
چکیده
Since Artificial Intelligence (AI) software uses techniques like deep lookahead search and stochastic optimization of huge neural networks to fit mammoth datasets, it often results in complex behavior that is difficult for people to understand. Yet organizations are deploying AI algorithms in many mission-critical settings. In order to trust their behavior, we must make it intelligible — either by using inherently interpretable models or by developing methods for explaining otherwise overwhelmingly complex decisions by local approximation, vocabulary alignment, and interactive dialog.
منابع مشابه
Intelligible Messaging – Activity-oriented Instant Messaging
I-X Process Panels are used to support users who are carrying out processes and responding to events in a cooperative working environment. The panels support the tracking of personal or group issues, the planning and execution of activities and the checking of constraints. Panels can be connected to other panels, and also to a range of services, agents and other cooperative working support tool...
متن کاملImage2speech: Automatically generating audio descriptions of images
This paper proposes a new task for artificial intelligence. The image2speech task generates a spoken description of an image. We present baseline experiments in which the neural net used is a sequence-to-sequence model with attention, and the speech synthesizer is clustergen. Speech is generated from four different types of segmentations: two that require a language with known orthography (word...
متن کاملThe Expressive Stance: Intentionality, Expression, and Machine Art
This paper proposes a new interpretive stance for interpreting artistic works and performances that is relevant to artificial intelligence research but also has broader implications. Termed the expressive stance, this stance makes intelligible a critical distinction between present-day machine art and human art, but allows for the possibility that future machine art could find a place alongside...
متن کاملPhase Transitions in Machine Learning
Phase transitions typically occur in combinatorial computational problems and have important consequences, especially with the current spread of statistical relational learning and of sequence learning methodologies. In Phase Transitions in Machine Learning the authors begin by describing in detail this phenomenon and the extensive experimental investigation that supports its presence. They the...
متن کاملBANTER: a Bayesian network tutoring shell
We present an educational tool for bringing the information contained in a Bayesian network to the end user in an easily intelligible form. The BANTER shell is designed to tutor users in evaluation of hypotheses and selection of optimal diagnostic procedures. BANTER can be used with any Bayesian network containing nodes that can be classified into hypotheses, observations, and diagnostic proced...
متن کامل