Impact of On-chip Interconnect on In-memory Acceleration of Deep Neural Networks

With the widespread use of Deep Neural Networks (DNNs), machine learning algorithms have evolved in two diverse directions—one with ever-increasing connection density for better accuracy and the other with more compact sizing for energy efficiency. The increase in connection density increases on-chip data movement, which makes efficient on-chip communication a critical function of the DNN accelerator. The contribution of this work is threefold. First, we illustrate that the point-to-point (P2P)-based interconnect is incapable of handling a high volume of on-chip data movement for DNNs. Second, we evaluate P2P and network-on-chip (NoC) interconnect (with a regular topology such as a mesh) for SRAM- and ReRAM-based in-memory computing (IMC) architectures for a range of DNNs. This analysis shows the necessity for the optimal interconnect choice for an IMC DNN accelerator. Finally, we perform an experimental evaluation for different DNNs to empirically obtain the performance of the IMC architecture with both NoC-tree and NoC-mesh. We conclude that, at the tile level, NoC-tree is appropriate for compact DNNs employed at the edge, and NoC-mesh is necessary to accelerate DNNs with high connection density. Furthermore, we propose a technique to determine the optimal choice of interconnect for any given DNN. In this technique, we use analytical models of NoC to evaluate end-to-end communication latency of any given DNN. We demonstrate that the interconnect optimization in the IMC architecture results in up to 6 × improvement in energy-delay-area product for VGG-19 inference compared to the state-of-the-art ReRAM-based IMC architectures.

Misalignment fading effects on the ACC performance of relay-assisted MIMO/FSO systems over atmospheric turbulence channels

<span>The continuous development of internet of things (IoT) technology enables many devices to be interconnected through the external environment. Meanwhile, 5G technology provides an enhanced quality of services with high data transmission rates, requiring IoT implementation in the 5G architecture. Free-space optical communication (FSO) is considered a promising technique that can provide high-speed communication links, so FSO is an optimal choice for wireless networks to fulfill the full potential of 5G technology, providing speeds of 100 Gb/s or more. By implementing 5G features in IoT, IoT coverage and performance will be enhanced by using FSO models. Therefore, the paper proposed and investigated the multiple-input and multiple-output/free-space optical communication (MIMO/FSO) model using subcarrier quadrature amplitude modulation (SC-QAM) and relay stations over atmospheric turbulence channels by log-normal and gamma-gamma distribution under different turbulence conditions. The performance is examined based on the average channel capacity (ACC), which is expressed in terms of average spectral efficiency (ASE) parameters while changing the different parameters of the model. The mathematical formulas of ACC for atmospheric turbulence cases are calculated and discussed the influence of turbulence strength, the different number of relay stations, misalignment effects, and different MIMO configurations.</span>

Optimal location of resources and Steiner symmetry in a population dynamics model in heterogeneous environments

The subject of this paper is inspired by Cantrell and Cosner (1989) and Cosner, Cuccu and Porru (2013). Cantrell and Cosner (1989) investigate the dynamics of a population in heterogeneous environments by means of diffusive logistic equations. An important part of their study consists in finding sufficient conditions which guarantee the survival of the species. Mathematically, this task leads to the weighted eigenvalue problem \(-\Delta u =\lambda m u \) in a bounded smooth domain \(\Omega\subset \mathbb{R}^N\), \(N\geq 1\), under homogeneous Dirichlet boundary conditions, where \(\lambda \in \mathbb{R}\) and \(m\in L^\infty(\Omega)\). The domain \(\Omega\) represents the environment and \(m(x)\), called the local growth rate, says where the favourable and unfavourable habitats are located. Then, Cantrell and Cosner (1989) consider a class of weights \(m(x)\) corresponding to environments where the total sizes of favourable and unfavourable habitats are fixed, but their spatial arrangement is allowed to change; they determine the best choice among them for the population to survive. In our work we consider a sort of refinement of the result above. We write the weight \(m(x)\) as sum of two (or more) terms, i.e. \(m(x)=f_1(x)+f_2(x)\), where \(f_1(x)\) and \(f_2(x)\) represent the spatial densities of the two resources which contribute to form the local growth rate \(m(x)\). Then, we fix the total size of each resource allowing its spatial location to vary. As our first main result, we show that there exists an optimal choice of \(f_1(x)\) and \(f_2(x)\) and find the form of the optimizers. Our proof relies on some results in Cosner, Cuccu and Porru (2013) and on a new property (to our knowledge) about the classes of rearrangements of functions. Moreover, we show that if \(\Omega\) is Steiner symmetric, then the best arrangement of the resources inherits the same kind of symmetry. (Actually, this is proved in the more general context of the classes of rearrangements of measurable functions.

Current Progress in Breast Implant-Associated Anaplastic Large Cell Lymphoma

Breast implant-associated anaplastic large-cell lymphoma (BIA-ALCL) is an uncommon type of T-cell lymphoma. Although with a low incidence, the epidemiological data raised the biosafety and health concerns of breast reconstruction and breast augmentation for BIA-ALCL. Emerging evidence confirms that genetic features, bacterial contamination, chronic inflammation, and textured breast implant are the relevant factors leading to the development of BIA-ALCL. Almost all reported cases with a medical history involve breast implants with a textured surface, which reflects the role of implant surface characteristics in BIA-ALCL. With this review, we expect to highlight the most significant features on etiology, pathogenesis, diagnosis, and therapy of BIA-ALCL, as well as we review the physical characteristics of breast implants and their potential pathogenic effect and hopefully provide a foundation for optimal choice of type of implant with minimal morbidity.

Optimal Choice of Vaccination Scheduling in a Population Composed of Two Groups

In this chapter, the authors present a simple model to determine the optimal choice of vaccination scheduling for a society composed of two groups of individuals in order to minimize the economic loss only, assuming herd immunity. First, a simple classical SIR model is presented to form the basis of the analysis; second, the model is revised to include the effects of vaccination which in turn will be extended to include two heterogeneous groups of individuals forming a society. The solutions of relevant differential equations will then be used to calculate the total economic cost of each scenario presented.

Environmental Impact of EV Batteries and Their Recycling

Electrifying transportation is one of the biggest keys to solving the looming climate crisis. The demand for electric vehicles (EV) is booming in the last five years and will increase in the coming years. In this modern age, where EV is the finest means of transportation due to null exhaust gases, there is a dire need to think about ways of recycling and reusing those batteries associated with EVs. In this context, it is estimated that post-vehicle battery packs application will be crossed from 1.4 million to 6.8 million by the year 2035. Numerous researches have been done on the re-purposing and safe disposal of EV batteries. However, presently, Lithium-ion batteries (LiBs) are the optimal choice for electric transportation due to greater energy density, compact size, and extended life cycles. Nonetheless, the trade-off between re-purposing and disposal of LiBs is substantial for the protection of the environment and human health. Regrettably, Lithium-ion battery recycling percentage is only 3% currently whereas its revival is negligible.

Exploring the cooperation-cooperation mode between scenic spots and tourism enterprises under the dual-channel tourism supply chain

As the two most important nodes in the tourism supply chain, the competition, as well as the cooperation relationship between scenic spots and tourism enterprises, has a profound impact on the healthy development of the tourism industry. In this paper, we establish a dual-channel tourism supply chain model by making many assumptions about the operations of scenic spots and tourism enterprises. According to repeated game theory, we find that there is a certain trigger strategy for equilibrium strategy between scenic spots and tourism enterprises, and the cooperation-cooperation strategy is the optimal choice for scenic spots and tourism enterprises at this time.

AIC Algorithm for E-Wallet Payment in the COVID-19 Pandemic in Vietnam

The COVID-19 pandemic is continuously developing owing to the development of a new strain of coronavirus. Society is difficult with people's needs being higher, especially in daily spending and consumption, but many consumers are also tending to not want to keep too much cash on hand. In addition, the world is facing the COVID-19 pandemic, and Vietnam, in particular, has pushed businesses to focus on developing a cashless payment application called an e-wallet to pay bills to meet consumer needs. This chapter is based on concepts and theories of the technology acceptance model related to intention to use the e-wallets to form the research model comprising the independent factors of perceived usefulness, perceived ease of use, perceived risk, social influence, and perceived trust. This study uses the optimal choice of the AIC algorithm, and all the factors are acceptable.

Monetary incentives and fiscal policy mutual influence

The article considers the problems and practice of using the tools of monetary and fiscal stimulation of the economy. The main problems of the application of monetary instruments in practice in emerging markets are identified. The author paid special attention to the definition of classical monetary policy instruments and their role in economic growth in emerging markets. Critical assessment of the role of monetary policy instruments in stimulating economic growth is based on the practice of central banks in emerging markets. Recommendations for the analysis of the efficiency of monetary transmission are given. Problems of efficiency of application of fiscal stimulus instruments in emerging markets are raised. The mechanisms of the dependence between fiscal and monetary policies and the strengthening of such dependence in the case of a significant informal sector of the economy and an underdeveloped financial market are demonstrated. The author not only points out the need for coordination in choosing between monetary and fiscal policy but also advocates the idea of having an independent arbitrator between monetary and fiscal authorities in developing countries. The article also focuses on the analyzing algorithm of the use of monetary policy instruments for economic growth effectiveness. The conclusions made in the article will be especially useful for those who are interested in the issue of optimal choice between monetary and fiscal instruments to stimulate economic development in emerging markets.

Dynamic decision policy reconfiguration under outcome uncertainty

In uncertain or unstable environments, sometimes the best decision is to change your mind. To shed light on this flexibility, we evaluated how the underlying decision policy adapts when the most rewarding action changes. Human participants performed a dynamic two-armed bandit task that manipulated the certainty in relative reward (conflict) and the reliability of action-outcomes (volatility). Continuous estimates of conflict and volatility contributed to shifts in exploratory states by changing both the rate of evidence accumulation (drift rate) and the amount of evidence needed to make a decision (boundary height), respectively. At the trialwise level, following a switch in the optimal choice, the drift rate plummets and the boundary height weakly spikes, leading to a slow exploratory state. We find that the drift rate drives most of this response, with an unreliable contribution of boundary height across experiments. Surprisingly, we find no evidence that pupillary responses associated with decision policy changes. We conclude that humans show a stereotypical shift in their decision policies in response to environmental changes.