Search

The progress in the field of neural computation hinges on the use of hardware more efficient than the conventional microprocessors. Recent works have shown that mixed-signal integrated memristive circuits, especially their passive ('0T1R') variety, may increase the neuromorphic network performance dramatically, leaving far behind their digital counterparts. The major obstacle, however, is relatively immature memristor technology so that only limited functionality has been demonstrated to date. Here we experimentally demonstrate operation of one-hidden layer perceptron classifier entirely in the mixed-signal integrated hardware, comprised of two passive 20x20 metal-oxide memristive crossbar arrays, board-integrated with discrete CMOS components. The demonstrated multilayer perceptron network, whose complexity is almost 10x higher as compared to previously reported functional neuromorphic classifiers based on passive memristive circuits, achieves classification fidelity within 3 percent of that obtained in simulations, when using ex-situ training approach. The successful demonstration was facilitated by improvements in fabrication technology of memristors, specifically by lowering variations in their I-V characteristics., Comment: 26 pages, 12 figures

We experimentally demonstrate classification of 4x4 binary images into 4 classes, using a 3-layer mixed-signal neuromorphic network ("MLP perceptron"), based on two passive 20x20 memristive crossbar arrays, board-integrated with discrete CMOS components. The network features 10 hidden-layer and 4 output-layer analog CMOS neurons and 428 metal-oxide memristors, i.e. is almost an order of magnitude more complex than any previously reported functional memristor circuit. Moreover, the inference operation of this classifier is performed entirely in the integrated hardware. To deal with larger crossbar arrays, we have developed a semi-automatic approach to their forming and testing, and compared several memristor training schemes for coping with imperfect behavior of these devices, as well as with variability of analog CMOS neurons. The effectiveness of the proposed schemes for defect and variation tolerance was verified experimentally using the implemented network and, additionally, by modeling the operation of a larger network, with 300 hidden-layer neurons, on the MNIST benchmark. Finally, we propose a simple modification of the implemented memristor-based vector-by-matrix multiplier to allow its operation in a wider temperature range., Comment: 4 pages, 13 figures

We have fabricated and successfully tested an analog vector-by-matrix multiplier, based on redesigned 10x12 arrays of 55 nm commercial NOR flash memory cells. The modified arrays enable high-precision individual analog tuning of each cell, with sub-1% accuracy, while keeping the highly optimized cells, with their long-term state retention, intact. The array has an area of 0.33 um^2 per cell, and is at least one order of magnitude more dense than the reported prior implementations of nonvolatile analog memories. The demonstrated vector-by-vector multiplier, using gate coupling to additional periphery cells, has ~2% precision, limited by the aggregate effect of cell noise, retention, mismatch, process variations, tuning precision, and capacitive crosstalk. A differential version of the multiplier has allowed us to demonstrate sub-3% temperature drift of the output signal in the range between 25C and 85C., Comment: 4 pages, 11 pages

We have designed, fabricated, and successfully tested a prototype mixed-signal, 28x28-binary-input, 10-output, 3-layer neuromorphic network ("MLP perceptron"). It is based on embedded nonvolatile floating-gate cell arrays redesigned from a commercial 180-nm NOR flash memory. The arrays allow precise (~1%) individual tuning of all memory cells, having long-term analog-level retention and low noise. Each array performs a very fast and energy-efficient analog vector-by-matrix multiplication, which is the bottleneck for signal propagation in most neuromorphic networks. All functional components of the prototype circuit, including 2 synaptic arrays with 101,780 floating-gate synaptic cells, 74 analog neurons, and the peripheral circuitry for weight adjustment and I/O operations, have a total area below 1 mm^2. Its testing on the common MNIST benchmark set (at this stage, with a relatively low weight import precision) has shown a classification fidelity of 94.65%, close to the 96.2% obtained in simulation. The classification of one pattern takes less than 1 us time and ~20 nJ energy - both numbers much better than for digital implementations of the same task. Estimates show that this performance may be further improved using a better neuron design and a more advanced memory technology, leading to a >10^2 advantage in speed and a >10^4 advantage in energy efficiency over the state-of-the-art purely digital (GPU and custom) circuits, at classification of large, complex patterns., Comment: 4 pages, 10 figures

We report on experimental demonstration of a mixed-signal 6-tap finite-impulse response (FIR) filter in which weights are implemented with titanium dioxide memristive devices. In the proposed design weight of a tap is stored with a relatively high precision in a memristive device that can be configured in field. Such approach enables efficient implementation of the most critical operation of an FIR filter, i.e. multiplication of the input signal with the tap weights and summation of the products from taps, in analog domain. As a result, the proposed design, when implemented with fully integrated hybrid CMOS/memristor circuit, is expected to be much more compact and energy efficient as compared to the state-of-the-art approaches., Comment: 4 pages

Metal-oxide memristors have emerged as promising candidates for hardware implementation of artificial synapses - the key components of high-performance, analog neuromorphic networks - due to their excellent scaling prospects. Since some advanced cognitive tasks require spiking neuromorphic networks, which explicitly model individual neural pulses (spikes) in biological neural systems, it is crucial for memristive synapses to support the spike-time-dependent plasticity (STDP), which is believed to be the primary mechanism of Hebbian adaptation. A major challenge for the STDP implementation is that, in contrast to some simplistic models of the plasticity, the elementary change of a synaptic weight in an artificial hardware synapse depends not only on the pre-synaptic and post-synaptic signals, but also on the initial weight (memristor's conductance) value. Here we experimentally demonstrate, for the first time, STDP protocols that ensure self-adaptation of the average memristor conductance, making the plasticity stable, i.e. insensitive to the initial state of the devices. The experiments have been carried out with 200-nm Al2O3/TiO2-x memristors integrated into 12x12 crossbars. The experimentally observed self-adaptive STDP behavior has been complemented with numerical modeling of weight dynamics in a simple system with a leaky-integrate-and-fire neuron with a random spike-train input, using a compact model of memristor plasticity, fitted for quantitatively correct description of our memristors., Comment: 13 pages, 5 figures

We have modified a commercial NOR flash memory array to enable high-precision tuning of individual floating-gate cells for analog computing applications. The modified array area per cell in a 180 nm process is about 1.5 um^2. While this area is approximately twice the original cell size, it is still at least an order of magnitude smaller than in the state-of-the-art analog circuit implementations. The new memory cell arrays have been successfully tested, in particular confirming that each cell may be automatically tuned, with ~1% precision, to any desired subthreshold readout current value within an almost three-orders-of-magnitude dynamic range, even using an unoptimized tuning algorithm. Preliminary results for a four-quadrant vector-by-matrix multiplier, implemented with the modified memory array gate-coupled with additional peripheral floating-gate transistors, show highly linear transfer characteristics over a broad range of input currents., Comment: 4 pages, 6 figures

This paper proposes a new numerical optimization algorithm inspired by the strawberry plant for solving complicated engineering problems. Plants like strawberry develop both runners and roots for propagation and search for water resources and minerals. In these plants, runners and roots can be thought of as tools for global and local searches, respectively. The proposed algorithm has three main differences with the trivial nature-inspired optimization algorithms: duplication-elimination of the computational agents at all iterations, subjecting all agents to both small and large movements from the beginning to end, and the lack of communication (information exchange) between agents. Moreover, it has the advantage of using only three parameters to be tuned by user. This algorithm is applied to standard test functions and the results are compared with GA and PSO. The proposed algorithm is also used to solve an open problem in the field of robust control theory. These simulations show that the proposed algorithm can very effectively solve complicated optimization problems.