RESUMEN
Despite impressive demonstrations of memristive behavior with halide perovskites, no clear pathway for material and device design exists for their applications in neuromorphic computing. Present approaches are limited to single element structures, fall behind in terms of switching reliability and scalability, and fail to map out the analog programming window of such devices. Here, we systematically design and evaluate robust pyridinium-templated one-dimensional halide perovskites as crossbar memristive materials for artificial neural networks. We compare two halide perovskite 1D inorganic lattices, namely (propyl)pyridinium and (benzyl)pyridinium lead iodide. The absence of conjugated, electron-rich substituents in PrPyr+ prevents edge-to-face type π-stacking, leading to enhanced electronic isolation of the 1D iodoplumbate chains in (PrPyr)[PbI3], and hence, superior resistive switching performance compared to (BnzPyr)[PbI3]. We report outstanding resistive switching behaviours in (PrPyr)[PbI3] on the largest flexible crossbar implementation (16 × 16) to date - on/off ratio (>105), long term retention (105 s) and high endurance (2000 cycles). Finally, we put forth a universal approach to comprehensively map the analog programming window of halide perovskite memristive devices - a critical prerequisite for weighted synaptic connections in artificial neural networks. This consequently facilitates the demonstration of accurate handwritten digit recognition from the MNIST database based on spike-timing-dependent plasticity of halide perovskite memristive synapses.
RESUMEN
Deep learning inference that needs to largely take place on the "edge" is a highly computational and memory intensive workload, making it intractable for low-power, embedded platforms such as mobile nodes and remote security applications. To address this challenge, this article proposes a real-time, hybrid neuromorphic framework for object tracking and classification using event-based cameras that possess desirable properties such as low-power consumption (5-14 mW) and high dynamic range (120 dB). Nonetheless, unlike traditional approaches of using event-by-event processing, this work uses a mixed frame and event approach to get energy savings with high performance. Using a frame-based region proposal method based on the density of foreground events, a hardware-friendly object tracking scheme is implemented using the apparent object velocity while tackling occlusion scenarios. The frame-based object track input is converted back to spikes for TrueNorth (TN) classification via the energy-efficient deep network (EEDN) pipeline. Using originally collected datasets, we train the TN model on the hardware track outputs, instead of using ground truth object locations as commonly done, and demonstrate the ability of our system to handle practical surveillance scenarios. As an alternative tracker paradigm, we also propose a continuous-time tracker with C ++ implementation where each event is processed individually, which better exploits the low latency and asynchronous nature of neuromorphic vision sensors. Subsequently, we extensively compare the proposed methodologies to state-of-the-art event-based and frame-based methods for object tracking and classification, and demonstrate the use case of our neuromorphic approach for real-time and embedded applications without sacrificing performance. Finally, we also showcase the efficacy of the proposed neuromorphic system to a standard RGB camera setup when simultaneously evaluated over several hours of traffic recordings.
RESUMEN
In this paper, we discuss the nonlinear computational power provided by dendrites in biological and artificial neurons. We start by briefly presenting biological evidence about the type of dendritic nonlinearities, respective plasticity rules and their effect on biological learning as assessed by computational models. Four major computational implications are identified as improved expressivity, more efficient use of resources, utilizing internal learning signals, and enabling continual learning. We then discuss examples of how dendritic computations have been used to solve real-world classification problems with performance reported on well known data sets used in machine learning. The works are categorized according to the three primary methods of plasticity used-structural plasticity, weight plasticity, or plasticity of synaptic delays. Finally, we show the recent trend of confluence between concepts of deep learning and dendritic computations and highlight some future research directions.
Asunto(s)
Dendritas , Modelos Neurológicos , Dendritas/fisiología , Aprendizaje Automático , Plasticidad Neuronal/fisiología , Neuronas/fisiologíaRESUMEN
The primary objective of this paper is to build classification models and strategies to identify breathing sound anomalies (wheeze, crackle) for automated diagnosis of respiratory and pulmonary diseases. In this work we propose a deep CNN-RNN model that classifies respiratory sounds based on Mel-spectrograms. We also implement a patient specific model tuning strategy that first screens respiratory patients and then builds patient specific classification models using limited patient data for reliable anomaly detection. Moreover, we devise a local log quantization strategy for model weights to reduce the memory footprint for deployment in memory constrained systems such as wearable devices. The proposed hybrid CNN-RNN model achieves a score of [Formula: see text] on four-class classification of breathing cycles for ICBHI'17 scientific challenge respiratory sound database. When the model is re-trained with patient specific data, it produces a score of [Formula: see text] for leave-one-out validation. The proposed weight quantization technique achieves ≈ 4 × reduction in total memory cost without loss of performance. The main contribution of the paper is as follows: Firstly, the proposed model is able to achieve state of the art score on the ICBHI'17 dataset. Secondly, deep learning models are shown to successfully learn domain specific knowledge when pre-trained with breathing data and produce significantly superior performance compared to generalized models. Finally, local log quantization of trained weights is shown to be able to reduce the memory requirement significantly. This type of patient-specific re-training strategy can be very useful in developing reliable long-term automated patient monitoring systems particularly in wearable healthcare solutions.
Asunto(s)
Redes Neurales de la Computación , Modelación Específica para el Paciente , Ruidos Respiratorios/clasificación , Procesamiento de Señales Asistido por Computador , HumanosRESUMEN
Shallow feed-forward networks are incapable of addressing complex tasks such as natural language processing that require learning of temporal signals. To address these requirements, we need deep neuromorphic architectures with recurrent connections such as deep recurrent neural networks. However, the training of such networks demand very high precision of weights, excellent conductance linearity and low write-noise- not satisfied by current memristive implementations. Inspired from optogenetics, here we report a neuromorphic computing platform comprised of photo-excitable neuristors capable of in-memory computations across 980 addressable states with a high signal-to-noise ratio of 77. The large linear dynamic range, low write noise and selective excitability allows high fidelity opto-electronic transfer of weights with a two-shot write scheme, while electrical in-memory inference provides energy efficiency. This method enables implementing a memristive deep recurrent neural network with twelve trainable layers with more than a million parameters to recognize spoken commands with >90% accuracy.
RESUMEN
This paper presents a real-time, low-complexity neuromorphic speech recognition system using a spiking silicon cochlea, a feature extraction module and a population encoding method based Neural Engineering Framework (NEF)/Extreme Learning Machine (ELM) classifier IC. Several feature extraction methods with varying memory and computational complexity are presented along with their corresponding classification accuracies. On the N-TIDIGITS18 dataset, we show that a fixed bin size based feature extraction method that votes across both time and spike count features can achieve an accuracy of 95% in software similar to previously report methods that use fixed number of bins per sample while using ~3× less energy and ~25× less memory for feature extraction (~1.5× less overall). Hardware measurements for the same topology show a slightly reduced accuracy of 94% that can be attributed to the extra correlations in hardware random weights. The hardware accuracy can be increased by further increasing the number of hidden nodes in ELM at the cost of memory and energy.
RESUMEN
Presence of wheezes in breathing sounds has been associated with several respiratory and pulmonary diseases. In this paper we present a novel low-complexity wheeze detection method based on frequency contour tracking for automatic wheeze detection. Two hardware friendly variants of the algorithm have also been proposed. Applying the proposed feature extraction algorithm we achieved very high classification accuracy (> 99%) at considerably low computational complexity (3×-6×) compared to earlier methods and the power consumption of the proposed method is shown to be significantly less (70×-100×) compared to `record and transmit' strategy in wearable devices.