Explaining Artificial Intelligence with Care
نویسندگان
چکیده
Abstract In the recent past, several popular failures of black box AI systems and regulatory requirements have increased research interest in explainable interpretable machine learning. Among different available approaches model explanation, partial dependence plots (PDP) represent one most famous methods for model-agnostic assessment a feature’s effect on response. Although PDPs are commonly used easy to apply they only provide simplified view thus risk be misleading. Relying interpretation given by PDP can dramatic consequences an application area such as forensics where decisions may directly affect people’s life. For this reason paper degree explainability is investigated real-world data set from field forensics: glass identification database. By means example aims illustrate two important aspects learning development practical point context (1) importance proper process selection, hyperparameter tuning validation well (2) careful artificial intelligence. purpose, concept extended multiclass classification problems e.g. data.
منابع مشابه
Artificial Intelligence for Artificial Artificial Intelligence
Crowdsourcing platforms such as Amazon Mechanical Turk have become popular for a wide variety of human intelligence tasks; however, quality control continues to be a significant challenge. Recently, we propose TURKONTROL, a theoretical model based on POMDPs to optimize iterative, crowdsourced workflows. However, they neither describe how to learn the model parameters, nor show its effectiveness...
متن کاملArtificial Intelligence with Uncertainty
Uncertainty exists widely in the subjective and objective world. In all kinds of uncertainty, randomness and fuzziness are the most important and fundamental. In this paper, the relationship between randomness and fuzziness is discussed. Uncertain states and their changes can be measured by entropy and hyper-entropy respectively. Taken advantage of entropy and hyper-entropy, the uncertainty of ...
متن کاملIntelligence in Artificial Intelligence
ions of these networks, eg, cube on cube (The Society of Mind by Marvin Minsky, MIT, 1985) and “learning” mechanisms (The Organization of Behavior by Donald O Hebb, 1949) may not constitute intelligence. These processes populate fields with values from the user environment which can be selectively used (per contra hard coded defined sets). For example, NEST Learning Thermostat uses input values...
متن کاملArtificial Intelligence in Healthcare
The article is a letter to the editor, so it has no abstract.
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Ki - Künstliche Intelligenz
سال: 2022
ISSN: ['1610-1987', '0933-1875']
DOI: https://doi.org/10.1007/s13218-022-00764-8