Enzo Fenoglio (Cisco)
Beyond Deep Learning – How to be ahead of tomorrow
Abstract: Deep Learning (DL) has demonstrated undeniable successes in object detection, language translation, video games, etc. However, DL is simply classical machine learning and statistical methods powered by GPU acceleration. DL networks only compute what they have been programmed or trained earlier, fail in a way that humans will never do, and are not able to learn from errors. The dangers represented by adversarial attacks create serious problems to self-driving vehicles or in cybersecurity, which can be partially recovered by Bayesian DL to model uncertainty. However, for the complex use cases we are confronted today, we need to move towards a Human-Level Artificial Intelligence (HLAI). HLAI is goal oriented, improves continuously from errors, provides explainability, and learns how-to-learn from few examples for acquiring a generalized knowledge to new situations. In this talk, I will present promising routes to overcome these challenges based on some advanced cognitive architectures.