"Adversarial Regularized Reconstruction for Anomaly Detection and Generation": joint work with Angelica Liguori, Ettore Ritacco and Francesco Sergio Pisani (ICAR CNR), accepted for presentation icdm2021. #outlierdetection #generativemodels #adversariallearning #icdm2021

Hi Twitter world, we are officially online at ismis2022.icar.cnr.it. This year’s conference theme spans over two main directions: “Towards Trustworthy Artificial Intelligence” and “Human-centric Artificial Intelligence”. 1/2

CyberSec4Europe's two independent external experts congratulating the owners of the top two project results. Want to follow more of today's events? Watch the live stream of Momentum! via our website. cybersec4europe.eu

amazon.it/avventure-del-… 😍

Yesterday I presented the paper "Cascade-based Echo-Chamber Detection" at the DEFirst Reading Group. Understanding how information propagates on social media is an important topic (see Pentagon fake news) and ProbML can be used to fit real data and study the phenomenon. Link👇

Pleased to share our accepted paper at ACM TIST! It delves into a key challenge in recommender systems-finding the best tradeoff between popularity bias and accuracy. #RecSys #MachineLearning #Bias

Ramona Procopio, ciao