Hunting the unknowns!

علیرضا وفایی صدر (پژوهشکده فیزیک پژوهشگاه دانش های بنیادی – IPM )

Abstract:  Anomaly/outlier detection is one challenging area in machine learning, especially for large datasets in high dimensions. How can one build a trap for a new type of animal if one knows nothing at all about that animal?

We have explored a general anomaly detection framework based on dimensionality reduction and unsupervised clustering (DRAMA). This approach identifies the primary prototypes in the data with anomalies detected by their large distances from the prototypes, either in the latent space or in the original, high-dimensional space. In this talk we will know more about the anomaly detection techniques and compare them to DRAMA in a wide variety of challenges. The talk is based on: https://arxiv.org/abs/1909.04060

یکشنبه 28 مهر  1398، ساعت 15:00

دانشکده فیزیک، طبقه اول کلاس فیزیک 3