Browse
Submit Data
Databases
API
Help

Dataset Information

21 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

A building block for hardware belief networks.

ABSTRACT: Belief networks represent a powerful approach to problems involving probabilistic inference, but much of the work in this area is software based utilizing standard deterministic hardware based on the transistor which provides the gain and directionality needed to interconnect billions of them into useful networks. This paper proposes a transistor like device that could provide an analogous building block for probabilistic networks. We present two proof-of-concept examples of belief networks, one reciprocal and one non-reciprocal, implemented using the proposed device which is simulated using experimentally benchmarked models.

SUBMITTER: Behin-Aein B

PROVIDER: S-EPMC4956763 | biostudies-literature | 2016 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Publications

A building block for hardware belief networks.

Behin-Aein Behtash B Diep Vinh V Datta Supriyo S

Scientific reports 20160721

Belief networks represent a powerful approach to problems involving probabilistic inference, but much of the work in this area is software based utilizing standard deterministic hardware based on the transistor which provides the gain and directionality needed to interconnect billions of them into useful networks. This paper proposes a transistor like device that could provide an analogous building block for probabilistic networks. We present two proof-of-concept examples of belief networks, one ...[more]

PMID: 27443521

Similar Datasets

Streaming Batch Eigenupdates for Hardware Neural Networks.

Project description:Neural networks based on nanodevices, such as metal oxide memristors, phase change memories, and flash memory cells, have generated considerable interest for their increased energy efficiency and density in comparison to graphics processing units (GPUs) and central processing units (CPUs). Though immense acceleration of the training process can be achieved by leveraging the fact that the time complexity of training does not scale with the network size, it is limited by the space complexity of stochastic gradient descent, which grows quadratically. The main objective of this work is to reduce this space complexity by using low-rank approximations of stochastic gradient descent. This low spatial complexity combined with streaming methods allows for significant reductions in memory and compute overhead, opening the door for improvements in area, time and energy efficiency of training. We refer to this algorithm and architecture to implement it as the streaming batch eigenupdate (SBE) approach.

| S-EPMC6691093 | biostudies-literature

Building block aspect ratio controls assembly, architecture, and mechanics of synthetic and natural protein networks.

Project description:Fibrous networks constructed from high aspect ratio protein building blocks are ubiquitous in nature. Despite this ubiquity, the functional advantage of such building blocks over globular proteins is not understood. To answer this question, we engineered hydrogel network building blocks with varying numbers of protein L domains to control the aspect ratio. The mechanical and structural properties of photochemically crosslinked protein L networks were then characterised using shear rheology and small angle neutron scattering. We show that aspect ratio is a crucial property that defines network architecture and mechanics, by shifting the formation from translationally diffusion dominated to rotationally diffusion dominated. Additionally, we demonstrate that a similar transition is observed in the model living system: fibrin blood clot networks. The functional advantages of this transition are increased mechanical strength and the rapid assembly of homogenous networks above a critical protein concentration, crucial for in vivo biological processes such as blood clotting. In addition, manipulating aspect ratio also provides a parameter in the design of future bio-mimetic and bio-inspired materials.

| S-EPMC10495373 | biostudies-literature

Hardware implementation of Bayesian network building blocks with stochastic spintronic devices

Project description:Bayesian networks are powerful statistical models to understand causal relationships in real-world probabilistic problems such as diagnosis, forecasting, computer vision, etc. For systems that involve complex causal dependencies among many variables, the complexity of the associated Bayesian networks become computationally intractable. As a result, direct hardware implementation of these networks is one promising approach to reducing power consumption and execution time. However, the few hardware implementations of Bayesian networks presented in literature rely on deterministic CMOS devices that are not efficient in representing the stochastic variables in a Bayesian network that encode the probability of occurrence of the associated event. This work presents an experimental demonstration of a Bayesian network building block implemented with inherently stochastic spintronic devices based on the natural physics of nanomagnets. These devices are based on nanomagnets with perpendicular magnetic anisotropy, initialized to their hard axes by the spin orbit torque from a heavy metal under-layer utilizing the giant spin Hall effect, enabling stochastic behavior. We construct an electrically interconnected network of two stochastic devices and manipulate the correlations between their states by changing connection weights and biases. By mapping given conditional probability tables to the circuit hardware, we demonstrate that any two node Bayesian networks can be implemented by our stochastic network. We then present the stochastic simulation of an example case of a four node Bayesian network using our proposed device, with parameters taken from the experiment. We view this work as a first step towards the large scale hardware implementation of Bayesian networks.

| S-EPMC7524796 | biostudies-literature

Hardware implementation of memristor-based artificial neural networks.

Project description:Artificial Intelligence (AI) is currently experiencing a bloom driven by deep learning (DL) techniques, which rely on networks of connected simple computing units operating in parallel. The low communication bandwidth between memory and processing units in conventional von Neumann machines does not support the requirements of emerging applications that rely extensively on large sets of data. More recent computing paradigms, such as high parallelization and near-memory computing, help alleviate the data communication bottleneck to some extent, but paradigm- shifting concepts are required. Memristors, a novel beyond-complementary metal-oxide-semiconductor (CMOS) technology, are a promising choice for memory devices due to their unique intrinsic device-level properties, enabling both storing and computing with a small, massively-parallel footprint at low power. Theoretically, this directly translates to a major boost in energy efficiency and computational throughput, but various practical challenges remain. In this work we review the latest efforts for achieving hardware-based memristive artificial neural networks (ANNs), describing with detail the working principia of each block and the different design alternatives with their own advantages and disadvantages, as well as the tools required for accurate estimation of performance metrics. Ultimately, we aim to provide a comprehensive protocol of the materials and methods involved in memristive neural networks to those aiming to start working in this field and the experts looking for a holistic approach.

| S-EPMC10912231 | biostudies-literature

Genetic dissection of the polyoxin building block-carbamoylpolyoxamic acid biosynthesis revealing the "pathway redundancy" in metabolic networks.

Project description:BACKGROUND: Polyoxin, a peptidyl nucleoside antibiotic, consists of three building blocks including a nucleoside skeleton, polyoximic acid (POIA), and carbamoylpolyoxamic acid (CPOAA), however, little is known about the "pathway redundancy" of the metabolic networks directing the CPOAA biosynthesis in the cell factories of the polyoxin producer. RESULTS: Here we report the genetic characterization of CPOAA biosynthesis with revealing a "pathway redundancy" in metabolic networks. Independent mutation of the four genes (polL-N and polP) directly resulted in the accumulation of polyoxin I, suggesting their positive roles for CPOAA biosynthesis. Moreover, the individual mutant of polN and polP also partially retains polyoxin production, suggesting the existence of the alternative homologs substituting their functional roles. CONCLUSIONS: It is unveiled that argA and argB in L-arginine biosynthetic pathway contributed to the "pathway redundancy", more interestingly, argB in S. cacaoi is indispensible for both polyoxin production and L-arginine biosynthesis. These data should provide an example for the research on the "pathway redundancy" in metabolic networks, and lay a solid foundation for targeted enhancement of polyoxin production with synthetic biology strategies.

| S-EPMC4029187 | biostudies-literature

Deep belief networks learn context dependent behavior.

Project description:With the goal of understanding behavioral mechanisms of generalization, we analyzed the ability of neural networks to generalize across context. We modeled a behavioral task where the correct responses to a set of specific sensory stimuli varied systematically across different contexts. The correct response depended on the stimulus (A,B,C,D) and context quadrant (1,2,3,4). The possible 16 stimulus-context combinations were associated with one of two responses (X,Y), one of which was correct for half of the combinations. The correct responses varied symmetrically across contexts. This allowed responses to previously unseen stimuli (probe stimuli) to be generalized from stimuli that had been presented previously. By testing the simulation on two or more stimuli that the network had never seen in a particular context, we could test whether the correct response on the novel stimuli could be generated based on knowledge of the correct responses in other contexts. We tested this generalization capability with a Deep Belief Network (DBN), Multi-Layer Perceptron (MLP) network, and the combination of a DBN with a linear perceptron (LP). Overall, the combination of the DBN and LP had the highest success rate for generalization.

| S-EPMC3966868 | biostudies-literature

Benchmarking Highly Parallel Hardware for Spiking Neural Networks in Robotics.

Project description:Animal brains still outperform even the most performant machines with significantly lower speed. Nonetheless, impressive progress has been made in robotics in the areas of vision, motion- and path planning in the last decades. Brain-inspired Spiking Neural Networks (SNN) and the parallel hardware necessary to exploit their full potential have promising features for robotic application. Besides the most obvious platform for deploying SNN, brain-inspired neuromorphic hardware, Graphical Processing Units (GPU) are well capable of parallel computing as well. Libraries for generating CUDA-optimized code, like GeNN and affordable embedded systems make them an attractive alternative due to their low price and availability. While a few performance tests exist, there has been a lack of benchmarks targeting robotic applications. We compare the performance of a neural Wavefront algorithm as a representative of use cases in robotics on different hardware suitable for running SNN simulations. The SNN used for this benchmark is modeled in the simulator-independent declarative language PyNN, which allows using the same model for different simulator backends. Our emphasis is the comparison between Nest, running on serial CPU, SpiNNaker, as a representative of neuromorphic hardware, and an implementation in GeNN. Beyond that, we also investigate the differences of GeNN deployed to different hardware. A comparison between the different simulators and hardware is performed with regard to total simulation time, average energy consumption per run, and the length of the resulting path. We hope that the insights gained about performance details of parallel hardware solutions contribute to developing more efficient SNN implementations for robotics.

| S-EPMC8275645 | biostudies-literature

What Is Central to Political Belief System Networks?

Project description:A central challenge for identifying core components of a belief system is examining the position of components within the structure of the entire belief system. We test whether operational (i.e., positions on issues) or symbolic (i.e., affective attachments to political groups and labels) components are most central by modeling a political belief system as a network of interconnected attitudes and beliefs. Across seven waves of representative panel data from New Zealand, we find that symbolic components are more central than operational components (ds range = 0.78-0.97). Symbolic components were also closer than operational components in the network to self-reported voting (d = -2.43), proenvironmental actions (ds = -1.71 and -1.63), and religious behaviors (d = -0.74). These findings are consistent with perspectives that emphasize the importance of symbolic politics in tying belief systems together and motivating behavior, and further the link between political belief system research and network science.

| S-EPMC6676336 | biostudies-literature

Asmic: An Exceptional Building Block for Isocyanide Alkylations.

Project description:Asmic addresses the long-standing challenge of alkylating isocyanides, providing access to isocyanides with diverse substitution patterns. The o-anisylsulfanyl group serves a critical dual role by facilitating deprotonation-alkylation and providing a latent nucleophilic site through an unusual arylsulfanyl-lithium exchange.

| S-EPMC6172042 | biostudies-literature

Use of a <i>cyclo</i>-P<sub>4</sub> building block - a way to networks of host-guest assemblies.

Project description:Despite the proven ability to form supramolecular assemblies via coordination to copper halides, organometallic building blocks based on four-membered cyclo-P4 ligands find only very rare application in supramolecular chemistry. To date, only three types of supramolecular aggregates were obtained based on the polyphosphorus end-deck complexes CpRTa(CO)2(η4-P4) (1a: CpR = Cp''; 1b: CpR = Cp'''), with none of them, however, possessing a guest-accessible void. To achieve this target, the use of silver salts of the weakly coordinating anion SbF6 - was investigated as to their self-assembly in the absence and in the presence of the template molecule P3Se4. The two-component self-assembly of the building block 1a and the coinage-metal salt AgSbF6 leads to the formation of 1D or 3D coordination polymers. However, when the template-driven self-assembly was attempted in the presence of an aliphatic dinitrile, the unprecedented barrel-like supramolecular host-guest assembly P3Se4@[{(Cp''Ta(CO)2(η4-P4))Ag}8]8+ of 2.49 nm in size was formed. Moreover, cyclo-P4-based supramolecules are connected in a 2D coordination network by dinitrile linkers. The obtained compounds were characterised by mass-spectrometry, 1H and 31P NMR spectroscopy and X-ray structure analysis.

| S-EPMC8163406 | biostudies-literature

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data