AAAI Workshop on Artificial Intelligence with Biased or Scarce Data (AIBSD) / / Kuan-Chuan Peng, Ziyan Wu

• Autore: Peng Kuan-Chuan
• Pubblicazione: Basel : , : MDPI - Multidisciplinary Digital Publishing Institute, , 2022
• Descrizione fisica: 1 electronic resource (186 p.)
• Disciplina: 006.3
• Soggetto topico: Technology: general issues; Artificial intelligence; History of engineering & technology
• Soggetto non controllato: permutation equivariance; optimization; gender bias; fairness; face-recognition models; facial attributes; social bias; bias detection; natural language processing; temporal bias; forecasting; contrastive learning; supervised contrastive learning; transfer learning; robustness; noisy labels; coresets; deep learning; contextualized embeddings; out-of-distribution generalization
• Persona (resp. second.): WuZiyan
• Nota di contenuto: About the Editors -- Statement of Peer Review -- Electricity Consumption Forecasting for Out-of-Distribution Time-of-Use Tariffs -- Measuring Embedded Human-Like Biases in Face Recognition Models -- Measuring Gender Bias in Contextualized Embeddings -- The Details Matter: Preventing Class Collapsein Supervised Contrastive Learning -- DAP-SDD: Distribution-Aware Pseudo Labeling for Small Defect Detection -- Quantifying Bias in a Face -- Verification System -- Super-Resolution for Brain MR Images from a Significantly Small Amount of Training Data -- Dual Complementary Prototype Learning for Few-Shot Segmentation -- Extracting Salient Facts from Company Reviews with Scarce Labels -- Long-Tail Zero and Few-Shot Learning via Contrastive Pretraining on and for Small Data -- Age Should Not Matter: -- Towards More Accurate Pedestrian Detection via Self-Training.
• Sommario/riassunto: This book is a collection of the accepted papers presented at the Workshop on Artificial Intelligence with Biased or Scarce Data (AIBSD) in conjunction with the 36th AAAI Conference on Artificial Intelligence 2022. During AIBSD 2022, the attendees addressed the existing issues of data bias and scarcity in Artificial Intelligence and discussed potential solutions in real-world scenarios. A set of papers presented at AIBSD 2022 is selected for further publication and included in this book.
• Altri titoli varianti: AAAI Workshop on Artificial Intelligence with Biased or Scarce Data
• Titolo autorizzato: AAAI Workshop on Artificial Intelligence with Biased or Scarce Data (AIBSD)
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 9910585937403321
• Lo trovi qui: Univ. Federico II