Feedforward networks in financial predictions: the future that modifies the present

The main goal of this paper is to show how relatively minor modifications of well-known algorithms (in particular, back propagation) can dramatically increase the performance of an artificial neural network (ANN) for time series prediction. We denote our proposed sets of modifications as the 'self-momentum', 'Freud' and 'Jung' rules. In our opinion, they provide an example of an alternative approach to the design of learning strategies for ANNs, one that focuses on basic mathematical conceptualization rather than on formalism and demonstration. The complexity of actual prediction problems makes it necessary to experiment with modelling possibilities whose inherent mathematical properties are often not well understood yet. The problem of time series prediction in stock markets is a case in point. It is well known that asset price dynamics in financial markets are difficult to trace, let alone to predict with an operationally interesting degree of accuracy. We therefore take financial prediction as a meaningful test bed for the validation of our techniques. We discuss in some detail both the theoretical underpinnings of the technique and our case study about financial prediction, finding encouraging evidence that supports the theoretical and operational viability of our new ANN specifications. Ours is clearly only a preliminary step. Further developments of ANN architectures with more and more sophisticated 'learning to learn' characteristics are now under study and test.     

Segmentation of news videos based on audio-video information

In this paper, we propose an innovative architecture to segment a news video into the so-called “stories” by both using the included video and audio information. Segmentation of news into stories is one of the key issues for achieving efficient treatment of news-based digital libraries. While the relevance of this research problem is widely recognized in the scientific community, we are in presence of a few established solutions in the field. In our approach, the segmentation is performed in two steps: first, shots are classified by combining three different anchor shot detection algorithms using video information only. Then, the shot classification is improved by using a novel anchor shot detection method based on features extracted from the audio track. Tests on a large database confirm that the proposed system outperforms each single video-based method as well as their combination.

A Novel Human Neutralizing mAb Recognizes Delta,Gamma and Omicron Variants of SARS-CoV-2 and Can Be Used in Combination with Sotrovimab

The dramatic experience with SARS-CoV-2 has alerted the scientific community to be ready to face new epidemics/pandemics caused by new variants. Among the therapies against the pandemic SARS-CoV-2 virus, monoclonal Antibodies (mAbs) targeting the Spike glycoprotein have represented good drugs to interfere in the Spike/ Angiotensin Converting Enzyme-2 (ACE-2) interaction, preventing virus cell entry and subsequent infection, especially in patients with a defective immune system. We obtained, by an innovative phage display selection strategy, specific binders recognizing different epitopes of Spike. The novel human antibodies specifically bind to Spike-Receptor Binding Domain (RBD) in a nanomolar range and interfere in the interaction of Spike with the ACE-2 receptor. We report here that one of these mAbs, named D3, shows neutralizing activity for virus infection in cell cultures by different SARS-CoV-2 variants and retains the ability to recognize the Omicron-derived recombinant RBD differently from the antibodies Casirivimab or Imdevimab. Since anti-Spike mAbs, used individually, might be unable to block the virus cell entry especially in the case of resistant variants, we investigated the possibility to combine D3 with the antibody in clinical use Sotrovimab, and we found that they recognize distinct epitopes and show additive inhibitory effects on the interaction of Omicron-RBD with ACE-2 receptor. Thus, we propose to exploit these mAbs in combinatorial treatments to enhance their potential for both diagnostic and therapeutic applications in the current and future pandemic waves of coronavirus.     

Efficient breadth first search on multi-GPU systems

Simple algorithms for the execution of a Breadth First Search on large graphs lead, running on clusters of GPUs, to a situation of load unbalance among threads and un-coalesced memory accesses, resulting in pretty low performances. To obtain a significant improvement on a single GPU and to scale by using multiple GPUs, we resort to a suitable combination of operations to rearrange data before processing them. We propose a novel technique for mapping threads to data that achieves a perfect load balance by leveraging prefix-sum and binary search operations. To reduce the communication overhead, we perform a pruning operation on the set of edges that needs to be exchanged at each BFS level. The result is an algorithm that exploits at its best the parallelism available on a single GPU and minimizes communication among GPUs. We show that a cluster of GPUs can efficiently perform a distributed BFS on graphs with billions of nodes.     

Cultural ‘demons’ as future builders

Usually, the shape of the future is seen as the result of a cultural flow that, according to some privileged cultural variable, like technology, goes undisturbed towards its own outcome. This is a quite naive attitude that has been very rarely successful. Both conventional technology and technology of the artificial show that, within culture, ‘demons’ are always active trying to exploit or even bypass standards in order to give birth to unexpected novelties. This is true within the pure technology area and even more in its artistic extensions. Therefore, no deterministic process is going on but, rather, a permanent remixing of chances that makes any prophecy impossible.     

Radiant floors integrated with PCM for indoor temperature control

The control of indoor thermal comfort in buildings through thermal inertia during the summer season plays a fundamental role in the design of energy efficient buildings, especially in the Mediterranean climate. In fact, lightweight, highly insulated buildings cannot provide the necessary mass to buffer thermal gains. Phase change materials (PCM) have been used to provide lightweight building components with the required thermal inertia without increasing their overall mass. So far the integration of PCM into lightweight piped radiant floors for the control of thermal comfort during summer cooling regimes has not been investigated. This paper reports the development of a lightweight piped radiant floor prototype with an integrated PCM layer aimed at buffering internal gains at constant temperature during summer cooling regimes without affecting its winter warming capacity. Both the construction of the laboratory specimens and the development of the optimized finite element models are detailed and the assessment of the floor performance in a simulated room is discussed.     

Deployment of RSS-Based Indoor Positioning Systems

Location estimation based on Received Signal Strength (RSS) is the prevalent method in indoor positioning. For such positioning systems, a massive collection of training samples is needed for their calibration. The accuracy of these methods is directly related to the placement of the reference points and the radio map used to compute the device location. Traditionally, deploying the reference points and building the radio map require human intervention and are extremely time-consuming. In this paper we present an approach to reduce the manual calibration efforts needed to deploy an RSS-based localization system, both when using only one RF technology or when using a combination of RF technologies. It is an automatic approach both to build a radio map in a given workspace by means of a signal propagation model, and to assess the system calibration that best fits the required accuracy by using a multi-objective genetic algorithm.     

1923 Gleno Dam Break: Case Study and Numerical Modeling

On the morning of December 1, 1923, the Gleno Dam (located in the Central Italian Alps) suddenly collapsed a few days after the first complete reservoir filling. Nearly 4.5×106??m3 of water was released. The consequent inundation caused significant destruction along the downstream valley and a death toll of at least 356 lives. This failure is the only historical case of dam break caused by structural deficiencies that has occurred in Italy. As a result, it has deeply influenced the evolution of Italian regulations regarding dam design and hydraulic risk evaluation. However, in spite of its relevance, this event has never been characterized from a hydraulic standpoint. This paper reports the main information obtained from the analysis of a vast amount of historical documents regarding the Gleno Dam break to set up a case study useful for validating dam-break models in mountain settings. Moreover, it presents the main results of one-dimensional (1D) modeling of the dam break wave propagation accomplished with a first-order finite volume numerical scheme recently proposed in the literature for field applications. The overall effectiveness and reliability of the model are evaluated for this case characterized by very irregular topography. Finally, the practical relevance of several choices that the numerical reconstruction of this kind of event demands is tested.