Search

Geometrical interpretation on U(1) gage theory of Dirac monopole, introduced here from the line integral\cite{Brandt} form of vector potentials, shows the gauge representation be multi-valued. In this paper, we construct Euclidean form of U(1) gauge theory on Dirac monopole based on the geometrical interpretation, derive the Dirac quantization condition geometrically. Even several theoretical methods which leads to Dirac quantization $eg=l/2$ has been developed, these are only possible under the postulate that gauge representation is single-valued. Using the line integration representation of vector potential, we derived the gauge equation which satisfies under the `phase gauge transformation' ($\mathbf{A}\to\mathbf{A}'=\mathbf{A}, \ \psi\to\psi'=e^{ie\Lambda}\psi$), also showed that the Dirac quantization condition can be derived by the geometrical boundary condition on this gauge equation, without any supposing of canonical quantization condition. We finally explain geometrical meaning of constant phase of gauge and how the Dirac quantization holds single valued gauge representation.

In recent years, the integration of large language models (LLMs) has revolutionized the field of robotics, enabling robots to communicate, understand, and reason with human-like proficiency. This paper explores the multifaceted impact of LLMs on robotics, addressing key challenges and opportunities for leveraging these models across various domains. By categorizing and analyzing LLM applications within core robotics elements -- communication, perception, planning, and control -- we aim to provide actionable insights for researchers seeking to integrate LLMs into their robotic systems. Our investigation focuses on LLMs developed post-GPT-3.5, primarily in text-based modalities while also considering multimodal approaches for perception and control. We offer comprehensive guidelines and examples for prompt engineering, facilitating beginners' access to LLM-based robotics solutions. Through tutorial-level examples and structured prompt construction, we illustrate how LLM-guided enhancements can be seamlessly integrated into robotics applications. This survey serves as a roadmap for researchers navigating the evolving landscape of LLM-driven robotics, offering a comprehensive overview and practical guidance for harnessing the power of language models in robotics development., Comment: 24 pages, 5 figures, Published in Intelligent Service Robotics (ISR)

The MFEM (Modular Finite Element Methods) library is a high-performance C++ library for finite element discretizations. MFEM supports numerous types of finite element methods and is the discretization engine powering many computational physics and engineering applications across a number of domains. This paper describes some of the recent research and development in MFEM, focusing on performance portability across leadership-class supercomputing facilities, including exascale supercomputers, as well as new capabilities and functionality, enabling a wider range of applications. Much of this work was undertaken as part of the Department of Energy's Exascale Computing Project (ECP) in collaboration with the Center for Efficient Exascale Discretizations (CEED)., Comment: 19 pages, 19 figures

We aim to solve the problem of spatially localizing composite instructions referring to space: space grounding. Compared to current instance grounding, space grounding is challenging due to the ill-posedness of identifying locations referred to by discrete expressions and the compositional ambiguity of referring expressions. Therefore, we propose a novel probabilistic space-grounding methodology (LINGO-Space) that accurately identifies a probabilistic distribution of space being referred to and incrementally updates it, given subsequent referring expressions leveraging configurable polar distributions. Our evaluations show that the estimation using polar distributions enables a robot to ground locations successfully through $20$ table-top manipulation benchmark tests. We also show that updating the distribution helps the grounding method accurately narrow the referring space. We finally demonstrate the robustness of the space grounding with simulated manipulation and real quadruped robot navigation tasks. Code and videos are available at https://lingo-space.github.io., Comment: Accepted by AAAI 2024

We introduce DynAMO, a reinforcement learning paradigm for Dynamic Anticipatory Mesh Optimization. Adaptive mesh refinement is an effective tool for optimizing computational cost and solution accuracy in numerical methods for partial differential equations. However, traditional adaptive mesh refinement approaches for time-dependent problems typically rely only on instantaneous error indicators to guide adaptivity. As a result, standard strategies often require frequent remeshing to maintain accuracy. In the DynAMO approach, multi-agent reinforcement learning is used to discover new local refinement policies that can anticipate and respond to future solution states by producing meshes that deliver more accurate solutions for longer time intervals. By applying DynAMO to discontinuous Galerkin methods for the linear advection and compressible Euler equations in two dimensions, we demonstrate that this new mesh refinement paradigm can outperform conventional threshold-based strategies while also generalizing to different mesh sizes, remeshing and simulation times, and initial conditions., Comment: 38 pages, 21 figures

Multi-band operation in wireless networks can improve data rates by leveraging the benefits of propagation in different frequency ranges. Distinctive beam management procedures in different bands complicate band assignment because they require considering not only the channel quality but also the associated beam management overhead. Reinforcement learning (RL) is a promising approach for multi-band operation as it enables the system to learn and adjust its behavior through environmental feedback. In this paper, we formulate a sequential decision problem to jointly perform band assignment and beam management. We propose a method based on hierarchical RL (HRL) to handle the complexity of the problem by separating the policies for band selection and beam management. We evaluate the proposed HRL-based algorithm on a realistic channel generated based on ray-tracing simulators. Our results show that the proposed approach outperforms traditional RL approaches in terms of reduced beam training overhead and increased data rates under a realistic vehicular channel.

The spoken language serves as an accessible and efficient interface, enabling non-experts and disabled users to interact with complex assistant robots. However, accurately grounding language utterances gives a significant challenge due to the acoustic variability in speakers' voices and environmental noise. In this work, we propose a novel speech-scene graph grounding network (SGGNet$^2$) that robustly grounds spoken utterances by leveraging the acoustic similarity between correctly recognized and misrecognized words obtained from automatic speech recognition (ASR) systems. To incorporate the acoustic similarity, we extend our previous grounding model, the scene-graph-based grounding network (SGGNet), with the ASR model from NVIDIA NeMo. We accomplish this by feeding the latent vector of speech pronunciations into the BERT-based grounding network within SGGNet. We evaluate the effectiveness of using latent vectors of speech commands in grounding through qualitative and quantitative studies. We also demonstrate the capability of SGGNet$^2$ in a speech-based navigation task using a real quadruped robot, RBQ-3, from Rainbow Robotics., Comment: 7 pages, 6 figures, Published at 2023 IEEE International Conference on Robot and Human Interactive Communication (RO-MAN), [Dohyun Kim, Yeseung Kim, Jaehwi Jang, and Minjae Song] contributed equally to this work

Cooperative relays improve reliability and coverage in wireless networks by providing multiple paths for data transmission. Relaying will play an essential role in vehicular networks at higher frequency bands, where mobility and frequent signal blockages cause link outages. To ensure connectivity in a relay-aided vehicular network, the relay selection policy should be designed to efficiently find unblocked relays. Inspired by recent advances in beam management in mobile millimeter wave (mmWave) networks, this paper address the question: how can the best relay be selected with minimal overhead from beam management? In this regard, we formulate a sequential decision problem to jointly optimize relay selection and beam management. We propose a joint relay selection and beam management policy based on deep reinforcement learning (DRL) using the Markov property of beam indices and beam measurements. The proposed DRL-based algorithm learns time-varying thresholds that adapt to the dynamic channel conditions and traffic patterns. Numerical experiments demonstrate that the proposed algorithm outperforms baselines without prior channel knowledge. Moreover, the DRL-based algorithm can maintain high spectral efficiency under fast-varying channels.

We introduce a higher simplicial generalization of the linear consensus model which shares several common features. The well-known linear consensus model is a gradient flow with a sum of squares of distances between each pair of points. Our newly suggested model is also represented as a gradient flow equipped with total $n$-dimensional volume functional consisting of $n+1$ points as a potential. In this manner, the linear consensus model coincides with the case of $n=1$ where distance is understood as the 1-dimensional volume. From a simple mathematical analysis, one can easily show that the linear consensus model (a gradient flow with 1-dimensional volume functional) collapses to one single point, which can be considered as a 0-complex. By extending this result, we show that a solution to our model converges to an $(n-1)$-dimensional affine subspace. We also perform several numerical simulations with an efficient algorithm that reduces the computational cost.

Despite the evolution of Convolutional Neural Networks (CNNs), their performance is surprisingly dependent on the choice of hyperparameters. However, it remains challenging to efficiently explore large hyperparameter search space due to the long training times of modern CNNs. Multi-fidelity optimization enables the exploration of more hyperparameter configurations given budget by early termination of unpromising configurations. However, it often results in selecting a sub-optimal configuration as training with the high-performing configuration typically converges slowly in an early phase. In this paper, we propose Multi-fidelity Optimization with a Recurring Learning rate (MORL) which incorporates CNNs' optimization process into multi-fidelity optimization. MORL alleviates the problem of slow-starter and achieves a more precise low-fidelity approximation. Our comprehensive experiments on general image classification, transfer learning, and semi-supervised learning demonstrate the effectiveness of MORL over other multi-fidelity optimization methods such as Successive Halving Algorithm (SHA) and Hyperband. Furthermore, it achieves significant performance improvements over hand-tuned hyperparameter configuration within a practical budget.

We provide the detailed asymptotic behavior for first-order aggregation models of heterogeneous oscillators. Due to the dissimilarity of natural frequencies, one could expect that all relative distances converge to definite positive value and furthermore that each oscillator converges to a possibly different stationary point. In order to establish the desired results, we introduce a novel method, called dimension reduction method that can be applied to a specific situation when the degree of freedom of the natural frequency is one. In this way, we would say that although a small perturbation is allowed, convergence toward an equilibrium of the gradient flow is still guaranteed. Several first-order aggregation models are provided as concrete examples by using the dimension reduction method to study the structure of the equilibrium, and numerical simulations are conducted to support theoretical results.

Cartoon domain has recently gained increasing popularity. Previous studies have attempted quality portrait stylization into the cartoon domain; however, this poses a great challenge since they have not properly addressed the critical constraints, such as requiring a large number of training images or the lack of support for abstract cartoon faces. Recently, a layer swapping method has been used for stylization requiring only a limited number of training images; however, its use cases are still narrow as it inherits the remaining issues. In this paper, we propose a novel method called Cross-domain Style mixing, which combines two latent codes from two different domains. Our method effectively stylizes faces into multiple cartoon characters at various face abstraction levels using only a single generator without even using a large number of training images.

The rapid increase in the adoption of Internet-of-Things (IoT) devices raises critical privacy concerns as these devices can access a variety of sensitive data. The current status quo of relying on manufacturers' cloud services to process this data is especially problematic since users cede control once their data leaves their home. Multiple recent incidents further call into question if vendors can indeed be trusted with users' data. At the same time, users desire compelling features supported by IoT devices and ML-based cloud inferences which compels them to subscribe to manufacturer-managed cloud services. An alternative to use a local in-home hub requires substantial hardware investment, management, and scalability limitations. This paper proposes Self-Serviced IoT (SSIoT), a clean-slate approach of using a hybrid hub-cloud setup to enable privacy-aware computation offload for IoT applications. Uniquely, SSIoT enables opportunistic computation offload to public cloud providers while still ensuring that the end-user retains complete end-to-end control of their private data reducing the trust required from public cloud providers. We show that SSIoT can leverage emerging function-as-a-service computation (e.g. AWS Lambda) to make these offloads cost-efficient, scalable and high performance as long as key limitations of being stateless, limited resources, and security isolation can be addressed. We build an end-to-end prototype of SSIoT and evaluate it using several micro-benchmarks and example applications representing real-world IoT use cases. Our results show that SSIoT is highly scalable, as compared to local-only approaches which struggle with as little as 2-4 apps in parallel. We also show that SSIoT is cost-efficient (operating a smart doorbell for $10 a year) at the cost of minimal additional latency as compared to a local-only hub, even with a hardware ML accelerator.

We present a systematic algebraic approach for the weak coupling of Cauchy problems to multiple Lohe tensor models. For this, we identify an admissible Cauchy problem to the Lohe tensor (LT) model with a characteristic symbol consisting of four tuples in terms of a size vector, a natural frequency tensor, a coupling strength tensor and admissible initial configuration. In this way, the collection of all admissible Cauchy problems to the LT models is equivalent to the space of characteristic symbols. On the other hand, we introduce a binary operation, namely "{\it fusion operation}" as a binary operation between the characteristic symbols. It turns out that the fusion operation satisfies the associativity and admits the identity element in the space of characteristic symbols which naturally forms a monoid. By virtue of the fusion operation, the weakly coupled system of multiple LT models can be obtained by applying the fusion operation of multiple characteristic symbols corresponding to the LT models. As a concrete example, we consider a weak coupling of the swarm sphere model and the Lohe matrix model, and provide a sufficient framework leading to emergent dynamics to the proposed weakly coupled model.

In this paper, we review state-of-the-art results on the collective behaviors for Lohe type first-order aggregation models. Collective behaviors of classical and quantum many-body systems have received lots of attention from diverse scientific disciplines such as applied mathematics, control theory in engineering, nonlinear dynamics of statistical physics, etc. To model such collective dynamics, several phenomenological models were proposed in literature and their emergent dynamics were extensively studied in recent years. Among them, we present two Lohe type models: the Lohe tensor (LT) model and the Schrodinger-Lohe mode}, and present several sufficient conditions in unified frameworks via the Lyapunov functional approach for state diameters and dynamical systems theory approach for two-point correlation functions. We also present several numerical simulation results for the SL model.

In this paper, we design and analyze staggered discontinuous Galerkin methods of arbitrary polynomial orders for the stationary Navier-Stokes equations on polygonal meshes. The exact divergence-free condition for the velocity is satisfied without any postprocessing. The resulting method is pressure-robust so that the pressure approximation does not influence the velocity approximation. A new nonlinear convective term that earning non-negativity is proposed. The optimal convergence estimates for all the variables in $L^2$ norm are proved. Also, assuming that the rotational part of the forcing term is small enough, we are able to prove that the velocity error is independent of the Reynolds number and of the pressure. Furthermore, superconvergence can be achieved for velocity under a suitable projection. Numerical experiments are provided to validate the theoretical findings and demonstrate the performances of the proposed method.

A tensor is a multi-dimensional array of complex numbers, and the Lohe tensor model is an aggregation model on the space of tensors with the same rank and size. It incorporates previously well-studied aggregation models on the space of low-rank tensors such as the Kuramoto model, Lohe sphere and matrix models as special cases. Due to its structural complexities in cubic interactions for the Lohe tensor model, explicit construction of solutions with specific structures looks daunting. Recently, we obtained completely separable states by associating rank-1 tensors. In this paper, we further investigate another type of solutions, namely "{\it quadratically separable states}" consisting of tensor products of matrices and their component rank-2 tensors are solutions to the double matrix model whose emergent dynamics can be studied using the same methodology of the Lohe matrix model.

We study emergent asymptotic dynamics for the first and second-order high-dimensional Kuramoto models on Stiefel manifolds which extend the previous consensus models on Riemannian manifolds including several matrix Lie groups. For the first-order consensus model on the Stiefel manifold proposed in [Markdahl et al, 2018], we show that the homogeneous ensemble relaxes the complete consensus state exponentially fast. On the other hand for a heterogeneous ensemble, we provide a sufficient condition leading to the phase-locked state in which relative distances between two states converge to definite values in a large coupling strength regime. We also propose a second-order extension of the first-order one by adding an inertial effect, and study emergent behaviors using Lyapunov functionals such as an energy functional and an averaged distance functional.

We demonstrate how to make the coordinate transformation or change of variables from Cartesian coordinates to curvilinear coordinates by making use of a convolution of a function with Dirac delta functions whose arguments are determined by the transformation functions between the two coordinate systems. By integrating out an original coordinate with a Dirac delta function, we replace the original coordinate with a new coordinate in a systematic way. A recursive use of Dirac delta functions allows the coordinate transformation successively. After replacing every original coordinate into a new curvilinear coordinate, we find that the resultant Jacobian of the corresponding coordinate transformation is automatically obtained in a completely algebraic way. In order to provide insights on this method, we present a few examples of evaluating the Jacobian explicitly without resort to the known general formula., Comment: 18 pages, no figures

We employ Dirac's bra-ket notation to define the inertia tensor operator that is independent of the choice of bases or coordinate system. The principal axes and the corresponding principal values for the elliptic plate are determined only based on the geometry. By making use of a general symmetric tensor operator, we develop a method of diagonalization that is convenient and intuitive in determining the eigenvector. We demonstrate that the bra-ket approach greatly simplifies the computation of the inertia tensor with an example of an $N$-dimensional ellipsoid. The exploitation of the bra-ket notation to compute the inertia tensor in classical mechanics should provide undergraduate students with a strong background necessary to deal with abstract quantum mechanical problems., Comment: 18 pages, 1 figure, Version published in J. Korean Phys. Soc

We present constants of motion for the finite-dimensional Lohe type aggregation models with frustration and we apply them to analyze the emergence of collective behaviors. The Lohe type models have been proposed as possible non-abelian and higher-dimensional generalizations of the Kuramoto model, which is a prototype phase model for synchronization. The aim of this paper is to study the emergent collective dynamics of these models under the effect of (interaction) frustration, which generalizes phase-shift frustrations in the Kuramoto model. To this end, we present constants of motion, i.e., conserved quantities along the flow generated by the models under consideration, and, from the perspective of the low-dimensional dynamics thus so obtained, derive several results concerning the emergent asymptotic patterns of the Kuramoto and Lohe sphere models.

Surface tension is a key parameter for understanding nucleation from the very initial stage of phase transformation. Although surface tension has been predicted to vary with the curvature of the liquid-vapor interface, particularly at the large curvature of, e.g., the subnanometric critical nucleus, experimental study still remains challenging due to inaccessibility to such a small cluster. Here, by directly measuring the critical size of a single capillary-condensed nanomeniscus using atomic force microscopy, we address the curvature dependence of surface tension of alcohols and observe the surface tension is doubled for ethanol and n-propanol with the radius-of-curvature of ~ -0.46 nm. We also find that the interface of larger negative (positive) curvature exhibits the larger (smaller) surface tension, which evidently governs nucleation at ~ 1 nm scale, indicating more facilitated nucleation than normally expected. Such well characterized curvature effects contribute to better understanding and accurate analysis of nucleation occurring in various fields including material science and atmospheric science., Comment: 5 pages, 4 figures and Supplementary Material(pdf)

We derive a general formula for the inertia tensor of a three-body system. By employing three independent Lagrange undetermined multipliers to express the vectors corresponding to the sides in terms of the position vectors of the vertices, we present the general covariant expression for the inertia tensor of the three particles of different masses. If $m_a/a=m_b/b=m_c/c=\rho$, then the center of mass coincides with the incenter of the triangle and the moment of inertia about the normal axis passing the center of mass is $I=\rho abc$, where $m_a$, $m_b$, and $m_c$ are the masses of the particles at $A$, $B$, and $C$, respectively, and $a$, $b$, and $c$ are the lengths of the line segments $\overline{BC}$, $\overline{CA}$, and $\overline{AB}$, respectively. The derivation and the corresponding results are closely related to the famous Heron's formula for the area of a triangle., Comment: 14 pages, 3 figures

In this paper, we present and analyze a staggered discontinuous Galerkin method for Darcy flows in fractured porous media on fairly general meshes. A staggered discontinuous Galerkin method and a standard conforming finite element method with appropriate inclusion of interface conditions are exploited for the bulk region and the fracture, respectively. Our current analysis weakens the usual assumption on the polygonal mesh, which can integrate more general meshes such as elements with arbitrarily small edges into our theoretical framework. We prove the optimal convergence estimates in $L^2$ error for all the variables by exploiting the Ritz projection. Importantly, our error estimates are shown to be fully robust with respect to the heterogeneity and anisotropy of the permeability coefficients. Several numerical experiments including meshes with small edges and anisotropic meshes are carried out to confirm the theoretical findings. Finally, our method is applied in the framework of unfitted mesh., Comment: 23 pages, 34 figures

A high-intensity proton beam exposure with 181 MeV energy has been conducted at Brookhaven Linac Isotope Producer facility on various material specimens for accelerator targetry applications, including titanium alloys as a beam window material. The radiation damage level of the analyzed capsule was 0.25 dpa at beam center region with an irradiation temperature around 120 degree C. Tensile tests showed increased hardness and a large decrease in ductility for the dual alpha+beta-phase Ti-6Al-4V Grade-5 and Grade-23 extra low interstitial alloys, with the near alpha-phase Ti-3Al-2.5V Grade-9 alloy still exhibiting uniform elongation of a few % after irradiation. Transmission Electron Microscope analyses on Ti-6Al-4V indicated clear evidence of a high-density of defect clusters with size less than 2 nm in each alpha-phase grain. The beta-phase grains did not contain any visible defects such as loops or black dots, while the diffraction patterns clearly indicated omega-phase precipitation in an advanced formation stage. The radiation-induced omega-phase transformation in the beta-phase could lead to greater loss of ductility in Ti-6Al-4V alloys in comparison with Ti-3Al-2.5V alloy with less beta-phase., Comment: 23 pages, 9 figures, In Press

This paper issues an integrated control system of self-driving autonomous vehicles based on the personal driving preference to provide personalized comfortable driving experience to autonomous vehicle users. We propose an Occupant's Preference Metric (OPM) which is defining a preferred lateral and longitudinal acceleration region with maximum allowable jerk for users. Moreover, we propose a vehicle controller based on control parameters enabling integrated lateral and longitudinal control via preference-aware maneuvering of autonomous vehicles. The proposed system not only provides the criteria for the occupant's driving preference, but also provides a personalized autonomous self-driving style like a human driver instead of a Robocar. The simulation and experimental results demonstrated that the proposed system can maneuver the self-driving vehicle like a human driver by tracking the specified criterion of admissible acceleration and jerk., Comment: 8 pages, 9 figures, NeurIPS 2019 Workshop: Machine Learning for Autonomous Driving (ML4AD)

Given the importance of understanding the interaction between mobile devices and their users, app usage patterns have been studied in various contexts. However, prior work has not fully investigated longitudinal changes to app usage behavior. In this paper, we present a longitudinal, large-scale study of mobile app usage based on a dataset collected from 162,006 iPhones and iPads over 4 years. We explore multiple dimensions of app usage pattern proving useful insights on how app usage changes over time. Our key findings include (i) app usage pattern changes over time both at the individual app level and the app category level (i.e. proportion of time a user spends using an app), (ii) users keep a small set of apps frequently launched (90% of iPhone users launch roughly 14-18 apps weekly), (iii) a small number of apps remain popular while some specific kinds of apps (e.g. Games) have a shorter life cycle compared to other apps of different categories. Finally, we discuss our findings and their implications, for example, a short-term study as an attempt to understand the general needs of mobile devices may not achieve useful results for the long term.

A high-strength dual alpha+beta phase titanium alloy Ti-6Al-4V is utilized as a material for beam windows in several accelerator target facilities. However, relatively little is known about how material properties of this alloy are affected by high-intensity proton beam irradiation. With plans to upgrade neutrino facilities at J-PARC and Fermilab to over 1 MW beam power, the radiation damage in the window material will reach a few displacements per atom (dpa) per year, significantly above the ~0.3 dpa level of existing data. The RaDIATE collaboration has conducted a high intensity proton beam irradiation of various target and window material specimens at BLIP facility, including a variety of titanium alloys. Post-Irradiation Examination of the specimens in the 1st capsule, irradiated at up to 0.25 dpa, is in progress. Tensile tests in a hot cell at PNNL exhibited a clear signature of radiation hardening and loss of ductility for Ti-6Al-4V, while Ti-3Al-2.5V, with less beta phase, exhibited less severe hardening. Microstructural investigations will follow to study the cause of the difference in tensile behavior between these alloys. High-cycle fatigue (HCF) performance is critical to the lifetime estimation of beam windows exposed to a periodic thermal stress from a pulsed proton beam. The 1st HCF data on irradiated titanium alloys are to be obtained by a conventional bend fatigue test at Fermilab and by an ultrasonic mesoscale fatigue test at Culham Laboratory. Specimens in the 2nd capsule, irradiated at up to ~1 dpa, cover typical titanium alloy grades, including possible radiation-resistant candidates. These systematic studies on the effects of radiation damage of titanium alloys are intended to enable us to predict realistic lifetimes of current beam windows made of Ti-6Al-4V and to extend the lifetime by choosing a more radiation and thermal shock tolerant alloy., Comment: 10 pages, 7 figures, In Proc. of IWSMT-14, 14th International Workshop on Spallation Materials Technology, 11th-17th Nov. 2018 at Fukushima, Japan

Computationally intensive Inference tasks of Deep neural networks have enforced revolution of new accelerator architecture to reduce power consumption as well as latency. The key figure of merit in hardware inference accelerators is the number of multiply-and-accumulation operations per watt (MACs/W), where, the state-of-the-arts MACs/W remains several hundreds Giga-MACs/W. We propose a Tera-MACS/W neural hardware inference Accelerator (TMA) with 8-bit activations and scalable integer weights less than 1-byte. The architectures main feature is configurable neural processing element for matrix-vector operations. The proposed neural processing element has Multiplier-less Massive Parallel Processor to work without any multiplications, which makes it attractive for energy efficient high-performance neural network applications. We benchmark our systems latency, power, and performance using Alexnet trained on ImageNet. Finally, we compared our accelerators throughput and power consumption to the prior works. The proposed accelerator outperforms the state of the art in terms of energy and area achieving 2.3 TMACS/W@1.0 V, 65 nm CMOS technology., Comment: 8 pages, 10 figures