### Top 10 Arxiv Papers Today in Emerging Technologies

##### #1. Importance of the window function choice for the predictive modelling of memristors
###### V. A. Slipko, Y. V. Pershin
Window functions are widely employed in memristor models to restrict the changes of the internal state variables to specified intervals. Here we show that the actual choice of window function is of significant importance for the predictive modelling of memristors. Using a recently formulated theory of memristor attractors, we demonstrate that whether attractor points exist depends on the type of window function used in the model. Our main findings are formulated in terms of two memristor attractor theorems, which apply to broad classes of memristor models. As an example of our findings, we predict the existence of attractor states in Biolek window function memristors and their absence in memristors described by the Joglekar window function, when such memristors are driven by periodic alternating polarity pulses. It is anticipated that the results of this study will contribute toward the development of more sophisticated models of memristive devices and systems.
more | pdf | html
###### Tweets
ComputerPapers: Importance of the window function choice for the predictive modelling of memristors. https://t.co/dTuGlAmuyL
None.
None.
###### Other stats
Sample Sizes : None.
Authors: 2
Total Words: 3570
Unqiue Words: 1108

##### #2. Perceptrons from Memristors
###### Francisco Silva, Mikel Sanz, João Seixas, Enrique Solano, Yasser Omar
Memristors, resistors with memory whose outputs depend on the history of their inputs, have been used with success in neuromorphic architectures, particularly as synapses or non-volatile memories. A neural network based on memristors could show advantages in terms of energy conservation and open up possibilities for other learning systems to be adapted to a memristor-based paradigm, both in the classical and quantum learning realms. No model for such a network has been proposed so far. Therefore, in order to fill this gap, we introduce models for single and multilayer perceptrons based on memristors. We adapt the delta rule to the memristor-based single-layer perceptron and the backpropagation algorithm to the memristor-based multilayer perceptron. We ran simulations of both the models and the training algorithms. These showed that both of them perform well and in accordance with Minsky-Papert's theorem, which motivates the possibility of building memristor-based hardware for a physical neural network.
more | pdf | html
###### Tweets
HNTweets: Perceptrons from memristors: https://t.co/ABIcvqkvXO Comments: https://t.co/QSb2T1bDt1
newsyc20: Perceptrons from memristors https://t.co/MQorZM276d (https://t.co/is1hMsBzCW)
deeplearning4j: Perceptrons from Memristors https://t.co/Kt28FcQFa9 #machinelearning
newsyc50: Perceptrons from memristors https://t.co/AVZsDEeTnw (https://t.co/OHMUPmBd1F)
jmsunico: sn-news: #electronics #ai #ml Perceptrons from Memristors https://t.co/Mp9MAWUmTu
hn_frontpage: Perceptrons from memristors L: https://t.co/4imYt7SJCQ C: https://t.co/f5wZ1iokTw
marcfawzi: In theory, this could replace GPUs for running neural networks, with far less energy and number of chip elements. This to GPU is like ASIC is to CPU, I think. https://t.co/aX0dK8HOqC #DeepLearning
nmfeeds: [O] https://t.co/DM3pyumCxW Perceptrons from Memristors. Memristors, resistors with memory whose outputs depend on the his...
nmfeeds: [NE] https://t.co/DM3pyumCxW Perceptrons from Memristors. Memristors, resistors with memory whose outputs depend on the hi...
reddit_ml: [1807.04912] Perceptrons from Memristors https://t.co/1xeU3Jb4Kz
FrontPageHN: Perceptrons from memristors https://t.co/BGBclE26z6 (cmts https://t.co/f1S58CLN28)
miketranchina: Perceptrons from Memristors Huge advance for artificial intelligence research https://t.co/SIKZE2kdwQ
McMurchie: The absolute next gen in AI that literally no one is talking about, not NeuralNetworks in software but NeuralNetworks in Hardware. The brain rewires itself, so should our computers. https://t.co/7Z5cA61cN7 https://t.co/yMz3zAlvFG
qmisanz: A nice contribution towards the design of neural networks based on memristors https://t.co/RmWyBZUR0b now, to the quantum case!
bootstrap_tokyo: memristorsからのPerceptrons https://t.co/wxbdbYdh6e
angsuman: Perceptrons from memristors https://t.co/VMst3Uz2rP
betterhn50: 51 – Perceptrons from memristors https://t.co/X1EiEDtOu9
HackerNewsPosts: Perceptrons from memristors Link: https://t.co/a6WMxRgsmm Cmts: https://t.co/eOqcBtfS6W
jreuben1: Perceptrons from Memristors https://t.co/xLX8zUhQJI
hackernews100: Perceptrons from memristors https://t.co/rXtCCEeLMX
betterhn100: Perceptrons from memristors https://t.co/kUoZgJY4KS
_7m26: Perceptrons from memristors https://t.co/WYYwbioHbM
hackernewsrobot: Perceptrons from memristors https://t.co/pkALVmiVng
ardenfreeman0: Perceptrons from memristors https://t.co/etvnNn4Glo #Perceptrons
AlexNogueiraMD: RT @qmisanz: A nice contribution towards the design of neural networks based on memristors https://t.co/RmWyBZUR0b now, to the quantum case!
qmisanz: RT @miketranchina: Perceptrons from Memristors Huge advance for artificial intelligence research https://t.co/SIKZE2kdwQ
None.
None.
###### Other stats
Sample Sizes : None.
Authors: 5
Total Words: 5313
Unqiue Words: 1697

##### #3. Code-division multiplexed resistive pulse sensor networks for spatio-temporal detection of particles in microfluidic devices
###### Ningquan Wang, Ruxiu Liu, Roozbeh Khodambashi, Norh Asmare, A. Fatih Sarioglu
Spatial separation of suspended particles based on contrast in their physical or chemical properties forms the basis of various biological assays performed on lab-on-achip devices. To electronically acquire this information, we have recently introduced a microfluidic sensing platform, called Microfluidic CODES, which combines the resistive pulse sensing with the code division multiple access in multiplexing a network of integrated electrical sensors. In this paper, we enhance the multiplexing capacity of the Microfluidic CODES by employing sensors that generate non-orthogonal code waveforms and a new decoding algorithm that combines machine learning techniques with minimum mean-squared error estimation. As a proof of principle, we fabricated a microfluidic device with a network of 10 code-multiplexed sensors and characterized it using cells suspended in phosphate buffer saline solution.
more | pdf | html
###### Tweets
nmfeeds: [O] https://t.co/NeNYftVbhg Code-division multiplexed resistive pulse sensor networks for spatio-temporal detection of par...
M157q_News_RSS: Code-division multiplexed resistive pulse sensor networks for spatio-temporal detection of particles in microfluidic https://t.co/LWfsgaqIrJ
None.
None.
###### Other stats
Sample Sizes : None.
Authors: 5
Total Words: 2792
Unqiue Words: 1028

##### #4. On-Chip Optical Convolutional Neural Networks
###### Hengameh Bagherian, Scott Skirlo, Yichen Shen, Huaiyu Meng, Vladimir Ceperic, Marin Soljacic
Convolutional Neural Networks (CNNs) are a class of Artificial Neural Networks(ANNs) that employ the method of convolving input images with filter-kernels for object recognition and classification purposes. In this paper, we propose a photonics circuit architecture which could consume a fraction of energy per inference compared with state of the art electronics.
more | pdf | html
###### Tweets
BrundageBot: On-Chip Optical Convolutional Neural Networks. Hengameh Bagherian, Scott Skirlo, Yichen Shen, Huaiyu Meng, Vladimir Ceperic, and Marin Soljacic https://t.co/T9ubJZcMRT
M157q_News_RSS: On-Chip Optical Convolutional Neural Networks. (arXiv:1808.03303v1 [https://t.co/65TXqFuFqK]) https://t.co/ArZoKrYRqD Convolutional Neural Networks (CNNs) are
nmfeeds: [O] https://t.co/7y5Xc9jrns On-Chip Optical Convolutional Neural Networks. Convolutional Neural Networks (CNNs) are a clas...
None.
None.
###### Other stats
Sample Sizes : None.
Authors: 6
Total Words: 7921
Unqiue Words: 2633

##### #5. Photonic Spiking Neural Networks - From Devices to Systems
###### Indranil Chakraborty, Gobinda Saha, Kaushik Roy
Spiking Neural Networks offer an event-driven and biologically realistic alternative to standard Artificial Neural Networks based on analog information processing which make them more suitable for energy-efficient hardware implementations of the functional units of the brain, namely, neurons and synapses. Despite extensive efforts to replicate the energy efficiency of the brain in standard von-Neumann architecture, the massive parallelism has remained elusive. This has led researchers to venture towards beyond von-Neumann computing or in-memory' computing paradigms based on CMOS and post-CMOS technologies. However, implementations of such platforms in the electrical domain faces potential limitations of switching speed and interconnect losses. Integrated Photonics offers an welcome alternative to standard microelectronic' platforms and have recently shown promise as a viable technology for spike-based neural processing systems. However, non-volatility has been identified as an important component of large-scale neuromorphic...
more | pdf | html
###### Tweets
nmfeeds: [O] https://t.co/GoTjRXPVcw Photonic Spiking Neural Networks - From Devices to Systems. Spiking Neural Networks offer an e...
DebasmitDas1: @karpathy Awesome work !!! I also want to share relevant work from our university : https://t.co/yJk2pA3BvE https://t.co/RUJ29WYTSg
DiveraAI: RT @DebasmitDas1: @karpathy Awesome work !!! I also want to share relevant work from our university : https://t.co/yJk2pA3BvE https://t.co/…
None.
None.
###### Other stats
Sample Sizes : None.
Authors: 3
Total Words: 8486
Unqiue Words: 2456

##### #6. Biological Optical-to-Chemical Signal Conversion Interface: A Small-scale Modulator for Molecular Communications
###### Laura Grebenstein, Jens Kirchner, Renata Stavracakis Peixoto, Wiebke Zimmermann, Florian Irnstorfer, Wayan Wicke, Arman Ahmadzadeh, Vahid Jamali, Georg Fischer, Robert Weigel, Andreas Burkovski, Robert Schober
Although many exciting applications of molecular communication (MC) systems are envisioned to be at microscale, the MC testbeds reported so far are mostly at macroscale. To link the macroworld to the microworld, we propose and demonstrate a biological signal conversion interface that can also be seen as a microscale modulator. In particular, the proposed interface transduces an optical signal, which is controlled using an LED, into a chemical signal by changing the pH of the environment. The modulator is realized using E. coli bacteria as microscale entity expressing the light-driven proton pump gloeorhodopsin from Gloeobacter violaceus. Upon inducing external light stimuli, these bacteria locally change their surrounding pH level by exporting protons into the environment. To verify the effectiveness of the proposed optical-to-chemical signal converter, we analyze the pH signal measured by a pH sensor, which serves as receiver. We develop an analytical parametric model for the induced chemical signal as a function of the applied...
more | pdf | html
None.
None.
###### Other stats
Sample Sizes : None.
Authors: 12
Total Words: 10597
Unqiue Words: 2819

##### #7. Neuro-memristive Circuits for Edge Computing: A review
###### Olga Krestinskaya, Alex Pappachen James, Leon O. Chua
The volume, veracity, variability and velocity of data produced from the ever increasing network of sensors connected to Internet pose challenges for power management, scalability and sustainability of cloud computing infrastructure. Increasing the data processing capability of edge computing devices at lower power requirements can reduce the overheads for cloud computing solutions. This paper provides the review of neuromorphic CMOS-memristive architectures that can be integrated into edge computing devices. We discuss why the neuromorphic architectures are useful for edge devices and show the advantages, drawbacks and open problems in the field of memristive circuit and architectures in terms of edge computing perspective.
more | pdf | html
###### Tweets
MachineLearn45: Neuro-memristive Circuits for Edge Computing: A review https://t.co/A1AWCbHabo https://t.co/onSydR1nOy
troysk704: Neuro-memristive Circuits for Edge Computing: A review https://t.co/ogTB8RRu1t
None.
None.
###### Other stats
Sample Sizes : None.
Authors: 3
Total Words: 14538
Unqiue Words: 3345

##### #8. Exploiting Inherent Error-Resiliency of Neuromorphic Computing to achieve Extreme Energy-Efficiency through Mixed-Signal Neurons
###### Baibhab Chatterjee, Priyadarshini Panda, Shovan Maity, Ayan Biswas, Kaushik Roy, Shreyas Sen
Neuromorphic computing, inspired by the brain, promises extreme efficiency for certain classes of learning tasks, such as classification and pattern recognition. The performance and power consumption of neuromorphic computing depends heavily on the choice of the neuron architecture. Digital neurons (Dig-N) are conventionally known to be accurate and efficient at high speed, while suffering from high leakage currents from a large number of transistors in a large design. On the other hand, analog/mixed-signal neurons are prone to noise, variability and mismatch, but can lead to extremely low-power designs. In this work, we will analyze, compare and contrast existing neuron architectures with a proposed mixed-signal neuron (MS-N) in terms of performance, power and noise, thereby demonstrating the applicability of the proposed mixed-signal neuron for achieving extreme energy-efficiency in neuromorphic computing. The proposed MS-N is implemented in 65 nm CMOS technology and exhibits > 100X better energy-efficiency across all...
more | pdf | html
None.
None.
###### Other stats
Sample Sizes : None.
Authors: 6
Total Words: 8735
Unqiue Words: 2531

##### #9. High-Accuracy and Fault Tolerant Stochastic Inner Product Design
###### Werner Haselmayr, Daniel Wiesinger, Michael Lunglmayr
In this work, we present a novel inner product design for stochastic computing. Stochastic computing is an emerging computing technique, that encodes a number in the probability of observing a one in a random bit stream. This leads to reduced hardware costs and high error tolerance. The proposed inner product design is based on a two-line bipolar encoding format and applies sequential processing of the input in a central accumulation unit. Sequential processing significantly increases the computation accuracy, since it allows for preliminary cancelation of carry bits. Moreover, the central accumulation unit gives a much better scalability compared to conventional adder tree approaches. We show that the proposed inner product design outperforms state-of-the-art designs in terms of hardware costs for high accuracy requirements and fault tolerance.
more | pdf | html
None.
None.
None.
###### Other stats
Sample Sizes : None.
Authors: 3
Total Words: 4134
Unqiue Words: 1127

##### #10. Pareto-Optimization Framework for Automated Network-on-Chip Design
###### Tzyy-Juin Kao, Wolfgang Fink
With the advent of multi-core processors, network-on-chip design has been key in addressing network performances, such as bandwidth, power consumption, and communication delays when dealing with on-chip communication between the increasing number of processor cores. As the numbers of cores increase, network design becomes more complex. Therefore, there is a critical need in soliciting computer aid in determining network configurations that afford optimal performance given resources and design constraints. We propose a Pareto-optimization framework that explores the space of possible network configurations to determine optimal network latencies, power consumption, and the corresponding link allocations. For a given number of routers, average network latency and power consumption as example performance objectives can be displayed in form of Pareto-optimal fronts, thus not only offering a design tool, but also enabling trade-off studies.
more | pdf | html
None.
None.
###### Other stats
Sample Sizes : None.
Authors: 2
Total Words: 6052
Unqiue Words: 2088

Assert is a website where the best academic papers on arXiv (computer science, math, physics), bioRxiv (biology), BITSS (reproducibility), EarthArXiv (earth science), engrXiv (engineering), LawArXiv (law), PsyArXiv (psychology), SocArXiv (social science), and SportRxiv (sport research) bubble to the top each day.

Papers are scored (in real-time) based on how verifiable they are (as determined by their Github repos) and how interesting they are (based on Twitter).

To see top papers, follow us on twitter @assertpub_ (arXiv), @assert_pub (bioRxiv), and @assertpub_dev (everything else).

To see beautiful figures extracted from papers, follow us on Instagram.

Tracking 58,338 papers.

###### Search
Sort results based on if they are interesting or reproducible.
Interesting
Reproducible
Online
###### Stats
Tracking 58,338 papers.