Search for: All records

Time-to-first-spike (TTFS) encoded spiking neural networks (SNNs), implemented using memristive crossbar arrays (MCA), achieve higher inference speed and energy efficiency compared to artificial neural networks (ANNs) and rate encoded SNNs. However, memristive crossbar arrays are vulnerable to conductance variations in the embedded memristor cells. These degrade the performance of TTFS encoded SNNs, namely their classification accuracy, with adverse impact on the yield of manufactured chips. To combat this yield loss, we propose a postmanufacture testing and tuning framework for these SNNs. In the testing phase, a timing encoded signature of the SNN, which is statistically correlated to the SNN performace, is extracted. In the tuning phase, this signature is mapped to optimal values of the tuning knobs (gain parameters), one parameter per layer, using a trained regressor, allowing very fast tuning (about 150ms). To further reduce the tuning overhead, we rank order hidden layer neurons based on their criticality and show that adding gain programmability only to 50% of the neurons is sufficient for performance recovery. Experiments show that the proposed framework can improve yield by up to 34% and average accuracy of memristive SNNs by up to 9%. 

Resistive random access Memory (RRAM) based spiking neural networks (SNN) are becoming increasingly attractive for pervasive energy-efficient classification tasks. However, such networks suffer from degradation of performance (as determined by classification accuracy) due to the effects of process variations on fabricated RRAM devices resulting in loss of manufacturing yield. To address such yield loss, a two-step approach is developed. First, an alternative test framework is used to predict the performance of fabricated RRAM based SNNs using the SNN response to a small subset of images from the test image dataset, called the SNN response signature (to minimize test cost). This diagnoses those SNNs that need to be performance-tuned for yield recovery. Next, SNN tuning is performed by modulating the spiking thresholds of the SNN neurons on a layer-by-layer basis using a trained regressor that maps the SNN response signature to the optimal spiking thresholdvalues during tuning. The optimal spiking threshold values are determined by an off-line optimization algorithm. Experiments show that the proposed framework can reduce the number of out-of-spec SNN devices by up to 54% and improve yield by as much as 8.6%. 

While resistive random access memory (RRAM) based deep neural networks (DNN) are important for low-power inference in IoT and edge applications, they are vulnerable to the effects of manufacturing process variations that degrade their performance (classification accuracy). However, to test the same post-manufacture, the (image) dataset used to train the associated machine learning applications may not be available to the RRAM crossbar manufacturer for privacy reasons. As such, the performance of DNNs needs to be assessed with carefully crafted dataset-agnostic synthetic test images that expose anomalies in the crossbar manufacturing process to the maximum extent possible. In this work, we propose a dataset-agnostic post-manufacture testing framework for RRAM-based DNNs using Entropy Guided Image Synthesis (EGIS). We first create a synthetic image dataset such that the DNN outputs corresponding to the synthetic images minimize an entropy-based loss metric. Next, a small subset (consisting of 10-20 images) of the synthetic image dataset, called the compact image dataset, is created to expedite testing. The response of the device under test (DUT) to the compact image dataset is passed to a machine learning based outlier detector for pass/fail labeling of the DUT. It is seen that the test accuracy using such synthetic test images is very close to that of contemporary test methods. 

Resistive random access memory (RRAM) based memristive crossbar arrays enable low power and low latency inference for convolutional neural networks (CNNs), making them suitable for deployment in IoT and edge devices. However, RRAM cells within a crossbar suffer from conductance variations, making RRAM-based CNNs vulnerable to degradation of their classification accuracy. To address this, the classification accuracy of RRAM based CNN chips can be estimated using predictive tests, where a trained regressor predicts the accuracy of a CNN chip from the CNN’s response to a compact test dataset. In this research, we present a framework for co-optimizing the pixels of the compact test dataset and the regressor. The novelty of the proposed approach lies in the ability to co-optimize individual image pixels, overcoming barriers posed by the computational complexity of optimizing the large numbers of pixels in an image using state-of-the-art techniques. The co-optimization problem is solved using a three step process: a greedy image downselection followed by backpropagation driven image optimization and regressor fine-tuning. Experiments show that the proposed test approach reduces the CNN classification accuracy prediction error by 31% compared to the state of the art. It is seen that a compact test dataset with only 2-4 images is needed for testing, making the scheme suitable for built-in test applications. 

Variability-induced accuracy degradation of RRAMbased DNNs is of great concern due to their significant potential for use in future energy-efficient machine learning architectures. To address this, we propose a two-step process. First, an enhanced testing procedure is used to predict DNN accuracy from a set of compact test stimuli (images). This test response (signature) is simply the concatenated vectors of output neurons of intermediate and final DNN layers over the compact test images applied. DNNs with a predicted accuracy below a threshold are then tuned based on this signature vector. Using a clustering based approach, the signature is mapped to the optimal tuning parameter values of the DNN (determined using off-line training of the DNN via backpropagation) in a single step, eliminating any post-manufacture training of the DNN weights (expensive). The tuning parameters themselves consist of the gains and offsets of the ReLU activation of neurons of the DNN on a per-layer basis and can be tuned digitally. Tuning is achieved in less than a second of tuning time, with yield improvements of over 45% with a modest accuracy reduction of 4% compared to digital DNNs. 

We report a mountain-scale record of erosion rates in the central Patagonian Andes from >10 million years (Ma) ago to present, which covers the transition from a fluvial to alpine glaciated landscape. Apatite (U-Th)/He ages of 72 granitic cobbles from alpine glacial deposits show slow erosion before ~6 Ma ago, followed by a two- to threefold increase in the spatially averaged erosion rate of the source region after the onset of alpine glaciations and a 15-fold increase in the top 25% of the distribution. This transition is followed by a pronounced decrease in erosion rates over the past ~3 Ma. We ascribe the pulse of fast erosion to local deepening and widening of valleys, which are characteristic features of alpine glaciated landscapes. The subsequent decline in local erosion rates may represent a return toward a balance between rock uplift and erosion. 

This paper presents a search for massive, charged, long-lived particles with the ATLAS detector at the Large Hadron Collider using an integrated luminosity of $$140~fb^{−1}$$ of proton-proton collisions at $$\sqrt{s}=13$$~TeV. These particles are expected to move significantly slower than the speed of light. In this paper, two signal regions provide complementary sensitivity. In one region, events are selected with at least one charged-particle track with high transverse momentum, large specific ionisation measured in the pixel detector, and time of flight to the hadronic calorimeter inconsistent with the speed of light. In the other region, events are selected with at least two tracks of opposite charge which both have a high transverse momentum and an anomalously large specific ionisation. The search is sensitive to particles with lifetimes greater than about 3 ns with masses ranging from 200 GeV to 3 TeV. The results are interpreted to set constraints on the supersymmetric pair production of long-lived R-hadrons, charginos and staus, with mass limits extending beyond those from previous searches in broad ranges of lifetime 

This report presents a comprehensive collection of searches for new physics performed by the ATLAS Collaboration during the Run~2 period of data taking at the Large Hadron Collider, from 2015 to 2018, corresponding to about 140~$$^{-1}$$ of $$\sqrt{s}=13$$~TeV proton--proton collision data. These searches cover a variety of beyond-the-standard model topics such as dark matter candidates, new vector bosons, hidden-sector particles, leptoquarks, or vector-like quarks, among others. Searches for supersymmetric particles or extended Higgs sectors are explicitly excluded as these are the subject of separate reports by the Collaboration. For each topic, the most relevant searches are described, focusing on their importance and sensitivity and, when appropriate, highlighting the experimental techniques employed. In addition to the description of each analysis, complementary searches are compared, and the overall sensitivity of the ATLAS experiment to each type of new physics is discussed. Summary plots and statistical combinations of multiple searches are included whenever possible.