1,721,038 research outputs found
Grandi eventi, Olimpiadi e comuni alpini: costruire un'eredità olimpica condivisa nelle vallate olimpiche
No full text
ColabNAS: Obtaining lightweight task-specific convolutional neural networks following Occam's razor
Full text linkThe current trend of applying transfer learning from convolutional neural networks (CNNs) trained on large datasets can be an overkill when the target application is a custom and delimited problem, with enough data to train a network from scratch. On the other hand, the training of custom and lighter CNNs requires expertise, in the from-scratch case, and or high-end resources, as in the case of hardware-aware neural architecture search (HW NAS), limiting access to the technology by non-habitual NN developers. For this reason, we present ColabNAS, an affordable HW NAS technique for producing lightweight task-specific CNNs. Its novel derivative-free search strategy, inspired by Occam's razor, allows to obtain state-of-the-art results on the Visual Wake Word dataset, a standard TinyML benchmark, in just 3.1 GPU hours using free online GPU services such as Google Colaboratory and Kaggle Kernel
A Miniature Direct-Drive Hydraulic Actuator for Wearable Haptic Devices based on Ferrofluid Magnetohydrodynamic Levitation
No full textHydraulic and pneumatic actuators in haptics offer the advantage of soft and compliant interfaces, with the drawback of cumbersome driving devices and limited modulation capabilities. We propose a miniature hydraulic actuator based on a linear electromagnetic motor with an embedded ferrofluid sealing. The solution has two main advantages: it shows no static friction due to the magnetohydrodynamic levitation effect of the ferrofluid, and the output force can be scaled (by varying the radius of the actuator) without introducing noise and friction of mechanical reduction mechanisms. Moreover, soft and compliant interfaces in the form of actuated pouches can be obtained on wearable devices with embedded actuators. As a concept prototype, we present here a compact and soft haptic thimble integrating the proposed actuator: experimental characterization at the bench, and perception experiments with the final prototype, evaluate the low-noise rendering capability of the method
Assessing the default risk by means of a discrete-time survival analysis approach
No full textIn this paper, the problem of company distress is assessed by means of a multi-period model that exploits the potentialities of the survival analysis approach when both survival times and regressors are measured at discrete points in time. The discrete-time hazards model can be used both as an empirical framework in the analysis of the causes of the deterioration process that leads to the default and as a tool for the prediction of the same event. Our results show that the prediction accuracy of the duration model is better than that provided by a single-period logistic model. It is also shown that the predictive power of the discrete-time survival analysis is enhanced when it is extended to allow for unobserved individual heterogeneity (frailty). Copyright 2008 John Wiley & Sons, Ltd
Human Recognition for Resource-Constrained Mobile Robot Applied to Covid-19 Disinfection
No full textThe global COVID-19 pandemic has stimulated the use of disinfection robots: in September 2021, following a European Commission’s action, 200 disinfection robots were delivered to European Hospitals. UV-C light is a common disinfection method, however, direct exposure to UV-C radiation is harmful and disinfection can be operated only in areas strictly forbidden to human personnel. We believe more advanced safety mechanisms are needed to increase the operational flexibility and safety level. We propose a safety mechanism based on vision and artificial intelligence, optimised for execution on mobile robot platforms. It analyses in real-time four video streaming and disables UV-C lamps when needed. Concerning other detection methods, it has a relatively wider and deeper range, and the capability to operate in a dynamic environment. We present the development of the method with a performance comparison of different implementation solutions, and an on-field evaluation through integration on a mobile disinfection robot
Default risk analysis via a discrete-time cure rate model
No full textCure models represent an appealing tool when analyzing default time data where two groups of companies are supposed to coexist: those which could eventually experience a default (uncured) and those which could not develop an endpoint (cured). One of their most interesting properties is the possibility to distinguish among covariates exerting their influence on the probability of belonging to the populations’ uncured fraction, from those affecting the default time distribution. This feature allows a separate analysis of the two dimensions of the default risk: whether the default can occur and when it will occur, given that it can occur. Basing our analysis on a large sample of Italian firms, the probability of being uncured is here estimated with a binary logit regression, whereas a discrete time version of a Cox’s proportional hazards approach is used to model the time distribution of defaults. The extension of the cure model as a forecasting framework is then accomplished by replacing the discrete time baseline function with an appropriate time-varying system level covariate, able to capture the underlying macroeconomic cycle. We propose a holdout sample procedure to test the classification power of the cure model. When compared with a single-period logit regression and a standard duration analysis approach, the cure model has proven to be more reliable in terms of the overall predictive performance
- …
