Ultra-low-resolution Infrared (IR) array sensors offer a low-cost, energy-efficient, and privacy-preserving solution for people counting, with applications such as occupancy monitoring. Previous work has shown that Deep Learning (DL) can yield superior performance on this task. However, the literature was missing an extensive comparative analysis of various efficient DL architectures for IR array-based people counting, that considers not only their accuracy, but also the cost of deploying them on memory- and energy-constrained Internet of Things (IoT) edge nodes. In this work, we address this need by comparing 6 different DL architectures on a novel dataset composed of IR images collected from a commercial 8x8 array, which we made openly available. With a wide architectural exploration of each model type, we obtain a rich set of Pareto-optimal solutions, spanning cross-validated balanced accuracy scores in the 55.70-82.70% range. When deployed on a commercial Microcontroller (MCU) by STMicroelectronics, the STM32L4A6ZG, these models occupy 0.41-9.28kB of memory, and require 1.10-7.74ms per inference, while consuming 17.18-120.43 $\mu$J of energy. Our models are significantly more accurate than a previous deterministic method (up to +39.9%), while being up to 3.53x faster and more energy efficient. Further, our models' accuracy is comparable to state-of-the-art DL solutions on similar resolution sensors, despite a much lower complexity. All our models enable continuous, real-time inference on a MCU-based IoT node, with years of autonomous operation without battery recharging.

Neural decoding involves correlating signals acquired from the brain to variables in the physical world like limb movement or robot control in Brain Machine Interfaces. In this context, this work starts from a specific pre-existing dataset of neural recordings from monkey motor cortex and presents a Deep Learning-based approach to the decoding of neural signals for grasp type classification. Specifically, we propose here an approach that exploits LSTM networks to classify time series containing neural data (i.e., spike trains) into classes representing the object being grasped. The main goal of the presented approach is to improve over state-of-the-art decoding accuracy without relying on any prior neuroscience knowledge, and leveraging only the capability of deep learning models to extract correlations from data. The paper presents the results achieved for the considered dataset and compares them with previous works on the same dataset, showing a significant improvement in classification accuracy, even if considering simulated real-time decoding.

With the increasing popularity of Internet of Things (IoT) devices, there is a growing need for energy-efficient Machine Learning (ML) models that can run on constrained edge nodes. Decision tree ensembles, such as Random Forests (RFs) and Gradient Boosting (GBTs), are particularly suited for this task, given their relatively low complexity compared to other alternatives. However, their inference time and energy costs are still significant for edge hardware. Given that said costs grow linearly with the ensemble size, this paper proposes the use of dynamic ensembles, that adjust the number of executed trees based both on a latency/energy target and on the complexity of the processed input, to trade-off computational cost and accuracy. We focus on deploying these algorithms on multi-core low-power IoT devices, designing a tool that automatically converts a Python ensemble into optimized C code, and exploring several optimizations that account for the available parallelism and memory hierarchy. We extensively benchmark both static and dynamic RFs and GBTs on three state-of-the-art IoT-relevant datasets, using an 8-core ultra-lowpower System-on-Chip (SoC), GAP8, as the target platform. Thanks to the proposed early-stopping mechanisms, we achieve an energy reduction of up to 37.9% with respect to static GBTs (8.82 uJ vs 14.20 uJ per inference) and 41.7% with respect to static RFs (2.86 uJ vs 4.90 uJ per inference), without losing accuracy compared to the static model.

We study asynchronous dynamics in a network of interacting agents updating their binary states according to a time-varying threshold rule. Specifically, agents revise their state asynchronously by comparing the weighted average of the current states of their neighbors in the interaction network with possibly heterogeneous time-varying threshold values. Such thresholds are determined by an exogenous signal representing an external influence field modeling the different agents' biases towards one state with respect to the other one. We prove necessary and sufficient conditions for global stability of consensus equilibria, i.e., equilibria where all agents have the same state, robustly with respect to the (constant or time-varying) external field. Our results apply to general weighted directed interaction networks and build on super-modularity properties of certain network coordination games whose best response dynamics coincide with the linear threshold dynamics. In particular, we introduce a novel notion of robust improvement paths for such games and characterize conditions for their existence.

Providing pedestrians and other vulnerable road users with a clear indication about a fully autonomous vehicle status and intentions is crucial to make them coexist. In the last few years, a variety of external interfaces have been proposed, leveraging different paradigms and technologies including vehicle-mounted devices (like LED panels), short-range on-road projections, and road infrastructure interfaces (e.g., special asphalts with embedded displays). These designs were experimented in different settings, using mockups, specially prepared vehicles, or virtual environments, with heterogeneous evaluation metrics. Promising interfaces based on Augmented Reality (AR) have been proposed too, but their usability and effectiveness have not been tested yet. This paper aims to complement such body of literature by presenting a comparison of state-of-the-art interfaces and new designs under common conditions. To this aim, an immersive Virtual Reality-based simulation was developed, recreating a well-known scenario represented by pedestrians crossing in urban environments under non-regulated conditions. A user study was then performed to investigate the various dimensions of vehicle-to-pedestrian interaction leveraging objective and subjective metrics. Even though no interface clearly stood out over all the considered dimensions, one of the AR designs achieved state-of-the-art results in terms of safety and trust, at the cost of higher cognitive effort and lower intuitiveness compared to LED panels showing anthropomorphic features. Together with rankings on the various dimensions, indications about advantages and drawbacks of the various alternatives that emerged from this study could provide important information for next developments in the field.

It is commonly understood that AI is one of the most disruptive technologies being developed. It may affect almost every aspect of society – from knowledge sharing to economic interactions, from making art crafts to finding cures for our diseases – and of personal life – from making friends, to finding a partner, from dealing with the pain for the loss of beloved people, to helping us managing our households through smart objects. Understanding the relationship between AI and society is a complex endeavour, since its shape and its evolution are not an immutable technological law, but instead the consequence of specific choices, both private and public, that could very well change over time and may of course influence its sustainability. Some powerful politicians like Vladimir Putin have declared that who will lead the researches in the field of AI, will lead the world, and of course many funds are coming from the armies (Harari, 2015) and from governments that seem to be working for monitoring and controlling us (Greenwald, 2015; Zuboff, 2018). Many others come from the finance world and are meant to increase the incomes of a few rich persons, regardless the risks ran by the rest of the population (O'Neil, 2016).

Alessandro Vespignani is a modern numen of computational epidemiology. He opened his talk with a brief survey on the history of numerical epidemic models, emphasizing how their advances virtually halted in the '50s. Those data-hungry models started to work egregiously only after the Big Data revolution: Multidimensional and granular data from a galaxy of public and private providers allowed epidemiologists to predict with striking accuracy the spreading of pandemics like H1N1, Ebola, and Zika. Inebriate by the power of Big Data, the scientific community explored new predictive models that were increasingly data-driven and less focused on understanding the underlying phenomena. That line of thinking led to glaring failures, epitomized by the infamous Google Flu Trends fiasco.

We show that discrete synaptic weights can be efficiently used for learning in large scale neural systems, and lead to unanticipated computational performance. We focus on the representative case of learning random patterns with binary synapses in single layer networks. The standard statistical analysis shows that this problem is exponentially dominated by isolated solutions that are extremely hard to find algorithmically. Here, we introduce a novel method that allows us to find analytical evidence for the existence of subdominant and extremely dense regions of solutions. Numerical experiments confirm these findings. We also show that the dense regions are surprisingly accessible by simple learning protocols, and that these synaptic configurations are robust to perturbations and generalize better than typical solutions. These outcomes extend to synapses with multiple states and to deeper neural architectures. The large deviation measure also suggests how to design novel algorithmic schemes for optimization based on local entropy maximization.