Alessandro Londei PhD

Sony Computer Science Laboratories Paris

Is the creativity process a prerogative of the human mind? Exploring new concepts, new different solutions to a given problem, or, more simply, taking into account an upcoming event never observed before, at present seems to be a challenging task for an artificial system. An efficient approach inspired by the human special way to address the concept of “new” should take into account a mix of hierarchical abstract and interconnected conceptual levels to be processed by an adaptively “fluid” artificial neural machine. Such approach would break several present constraints in the field of neural networks and deep learning, where the influence of static architectures and training algorithms limit the potential development of more promising neural topologies, mainly based on natural cognitive mechanisms, and allowing, at the same time, to deal with an incomplete knowledge of the perceived external world. In my research, I try to explore this unknown domain, looking for new neural architectures and efficient unsupervised training mechanisms driven by changing non-stationary environments, aimed at the identification and comprehension of an ecological artificial mind.

Ranking the effectiveness of worldwide COVID-19 government interventions

Assessing non-pharmaceutical interventions’ (NPIs) effectiveness to mitigate the spread of SARS-CoV-2 is critical to inform future preparedness response plans. We propose a modeling approach that combines four computational techniques merging statistical, inference, and artificial intelligence tools to evaluate the impact of NPIs on spreading the Covid19 pandemic. Our results indicate that a suitable combination of NPIs is necessary to curb the spread of the virus. Less disruptive and costly NPIs can be as effective as more intrusive, drastic ones (for example, a national lockdown). Using country-specific “what-if” scenarios, we assess how the effectiveness of NPIs depends on the local context, such as the timing of their adoption, opening the way for forecasting the effectiveness of future interventions.