From What to Why, the Growing Need for a Focus Shift Toward Explainability of AI in Digital Pathology

While it is impossible to deny the performance gains achieved through the incorporation of deep learning (DL) and other artificial intelligence (AI)-based techniques in pathology, minimal work has been done to answer the crucial question of why these algorithms predict what they predict. Tracing back classification decisions to specific input features allows for the quick identification of model bias as well as providing additional information toward understanding underlying biological mechanisms. In digital pathology, increasing the explainability of AI models would have the largest and most immediate impact for the image classification task. In this review, we detail some considerations that should be made in order to develop models with a focus on explainability.

On Outage Analysis in Overlay CCRN With RF Energy Harvesting and Co-Channel Interference

This paper explores the impact of co-channel interference (CCI) on the link outage and radio frequency (RF) energy harvesting (EH). For analysis, co-operative cognitive radio network (CCRN) architecture is considered as system model that supports one-way primary user (PU) and two-way secondary user (SU) communications, using an overlay mode of spectrum sharing. Closed form outage expressions are derived for both PU and SU network in presence of multiple antennas at PUs and CCI at SUs. The effect of CCI on the system performance is studied with respect to interference-to-noise-ratio (INR), transmission power, number of antennas and number of CCI sources. Performance gains are found to achieve ~ 20% and ~ 15% for PU and SU outage in two antenna system over a single antenna one.

The Role of Strategic Brightness and Entrepreneurship in Achieving High Performance Gains

The current research deals with everything related to activating the contemporary leadership style for business organizations according to their modern models (developmental, results, integrative, personal), as the style is chosen according to the leader's strategic repertoire and his inspirational powers in a specific position.As a result of the rapid developments and rapid changes in recent years and the intensification of competition in the markets as a whole, which made most organizations guarantee for themselves survival and sustainability in the markets and the results of their work achieved by the management approach to controlling their work and this is done according to the perspective of ingenuity or organizational creativity with the support of their thinkers.From here, the researcher identified the problem of the research with the question that confirms the extent to which it is possible to benefit from the level of creativity in the minds of the leaders of our local institutions, and what is their role in getting rid of routine obstacles that increase organizational deterioration to reach the pioneering performance in all their businesses. The aim of the current research is to analyze and explore the degree of strategic brightness in the commercial sector in Basra, to diagnose the relative importance of the components of strategic leadership in the environment of commercial activity and to try to analyze the level of awareness of the issue of creativity as an initiative. The business of modern Basra, namely )Al-Madda and Al-Taif companies(, amounting to (25) individuals out of a total of (35) individuals, at a rate of (71%). Descriptive and inferential tests will be conducted, the aim of which is to test the correlation and influence relationships between the variables of strategic brightness and investment leadership to gain the benefits of high performance and competitive advantage based on creativity and leadership, starting from leading and extending towards the endless competition in the market.

Optimized Cooling Power of a Vuilleumier Refrigerator with Limited Regeneration

Vuilleumier refrigerators provide cooling power by utilizing a heat source at temperatures above the ambient. This is particularly helpful in situations where waste heat is available and other power sources are limited. Vuilleumier refrigerators come in different technical configurations; here we analyze the thermodynamic performance of a configuration utilizing two displacer pistons with integrated regenerators. More specifically, we optimize the cooling power by optimizing the piston movement for a range of operation speeds. The optimization is based on the AS motion class for cyclic dynamics and uses an endoreversible model for the refrigerator. Our focus is on the influence of the regeneration extent present, and we find performance gains of about 17% for high regeneration extent and of about 28% for lower regeneration extent.

Optimal Matching in Julia Using Distributed Computing: Time Use Data (ATUS)

A brief tutorial on how to run optimal matching in Julia. The performance gains: it is twice faster than TraMineR on a dataset of about 20000 sequences with 96 steps each.

COVID-19 detection using cough sound analysis and deep learning algorithms

Reliable and rapid non-invasive testing has become essential for COVID-19 diagnosis and tracking statistics. Recent studies motivate the use of modern machine learning (ML) and deep learning (DL) tools that utilize features of coughing sounds for COVID-19 diagnosis. In this paper, we describe system designs that we developed for COVID-19 cough detection with the long-term objective of embedding them in a testing device. More specifically, we use log-mel spectrogram features extracted from the coughing audio signal and design a series of customized deep learning algorithms to develop fast and automated diagnosis tools for COVID-19 detection. We first explore the use of a deep neural network with fully connected layers. Additionally, we investigate prospects of efficient implementation by examining the impact on the detection performance by pruning the fully connected neural network based on the Lottery Ticket Hypothesis (LTH) optimization process. In general, pruned neural networks have been shown to provide similar performance gains to that of unpruned networks with reduced computational complexity in a variety of signal processing applications. Finally, we investigate the use of convolutional neural network architectures and in particular the VGG-13 architecture which we tune specifically for this application. Our results show that a unique ensembling of the VGG-13 architecture trained using a combination of binary cross entropy and focal losses with data augmentation significantly outperforms the fully connected networks and other recently proposed baselines on the DiCOVA 2021 COVID-19 cough audio dataset. Our customized VGG-13 model achieves an average validation AUROC of 82.23% and a test AUROC of 78.3% at a sensitivity of 80.49%.

A mental models approach for defining explainable artificial intelligence

Background Wide-ranging concerns exist regarding the use of black-box modelling methods in sensitive contexts such as healthcare. Despite performance gains and hype, uptake of artificial intelligence (AI) is hindered by these concerns. Explainable AI is thought to help alleviate these concerns. However, existing definitions for explainable are not forming a solid foundation for this work. Methods We critique recent reviews on the literature regarding: the agency of an AI within a team; mental models, especially as they apply to healthcare, and the practical aspects of their elicitation; and existing and current definitions of explainability, especially from the perspective of AI researchers. On the basis of this literature, we create a new definition of explainable, and supporting terms, providing definitions that can be objectively evaluated. Finally, we apply the new definition of explainable to three existing models, demonstrating how it can apply to previous research, and providing guidance for future research on the basis of this definition. Results Existing definitions of explanation are premised on global applicability and don’t address the question ‘understandable by whom?’. Eliciting mental models can be likened to creating explainable AI if one considers the AI as a member of a team. On this basis, we define explainability in terms of the context of the model, comprising the purpose, audience, and language of the model and explanation. As examples, this definition is applied to regression models, neural nets, and human mental models in operating-room teams. Conclusions Existing definitions of explanation have limitations for ensuring that the concerns for practical applications are resolved. Defining explainability in terms of the context of their application forces evaluations to be aligned with the practical goals of the model. Further, it will allow researchers to explicitly distinguish between explanations for technical and lay audiences, allowing different evaluations to be applied to each.

Organizing Entrepreneurial Teams: A Field Experiment on Autonomy over Choosing Teams and Ideas

Scholars have suggested that autonomy can lead to better entrepreneurial team performance. Yet, there are different types of autonomy, and they come at a cost. We shed light on whether two fundamental organizational design choices—granting teams autonomy to (1) choose project ideas to work on and (2) choose team members to work with—affect performance. We run a field experiment involving 939 students in a lean startup entrepreneurship course over 11 weeks. The aim is to disentangle the separate and joint effects of granting autonomy over choosing teams and choosing ideas compared with a baseline treatment with preassigned ideas and team members. We find that teams with autonomy over choosing either ideas or team members outperform teams in the baseline treatment as measured by pitch deck performance. The effect of choosing ideas is significantly stronger than the effect of choosing teams. However, the performance gains vanish for teams that are granted full autonomy over choosing both ideas and teams. This suggests the two forms of autonomy are substitutes. Causal mediation analysis reveals that the main effects of choosing ideas or teams can be partly explained by a better match of ideas with team members’ interests and prior network contacts among team members, respectively. Although homophily and lack of team diversity cannot explain the performance drop among teams with full autonomy, our results suggest that self-selected teams fall prey to overconfidence and complacency too early to fully exploit the potential of their chosen idea. We discuss the implications of these findings for research on organizational design, autonomy, and innovation.

Employment of Generalized Receiver for Group-Blind Intersymbol Detection in Downlink CDMA over Fading Multipath Channels

Group-blind multiuser detectors for uplink code-division multiple-access (CDMA) were developed by Wang and Host-Madsen. These detectors make use of the spreading sequences of known users to construct a group constraint to suppress the intracell interference. However, such techniques demand the estimation of the multipath channels and the delays of the known users. In the present paper, the blind generalized receiver is de-veloped for CDMA in fading multipath channels. The proposed generalized receiver utilizes the correlation in-formation between consecutively received signals to generate the corresponding group constraint. It is shown that by incorporating this group constraint, the proposed generalized receiver can provide different performance gains in both the uplink and downlink environments. Compared with the well-known group-blind detectors, our new methods only need to estimate the multipath channel of the desired user and do not require the channel es-timation of other users. Simulation results demonstrate that the proposed generalized receiver outperforms the conventional blind linear multiuser detectors.

Fuel Consumption Monitoring through COPERT Model—A Case Study for Urban Sustainability

Trackers installed in vehicles gives insights into many useful information and predict future mobility patterns and other aspects related to vehicles movement which can be used for smart and sustainable cities planning. A novel approach is used with the COPERT model to estimate fuel consumption on a huge dataset collected over a period of one year. Since the data size is enormous, Apache Spark, a big data analytical framework is used for performance gains while estimating vehicle fuel consumption with the lowest latency possible. The research presents peak and off-peak hours fuel consumption’s in three major cities, i.e., Karachi, Lahore and Islamabad. The results can assist smart city professionals to plan alternative trip routes, avoid traffic congestion in order to save fuel and time, and protect against urban pollution for effective smart city planning. The research will be a step towards Industry 5.0 by combining sustainable disruptive technologies.