Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userEnergy-efficient stochastic computing with superparamagnetic tunnel junctions
77
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Superparamagnetic tunnel junctions (SMTJs) have emerged as a competitive, realistic nanotechnology to support novel forms of stochastic computation in CMOS-compatible platforms. One of their applications is to generate random bitstreams suitable for use in stochastic computing implementations. We describe a method for digitally programmable bitstream generation based on pre-charge sense amplifiers. This generator is significantly more energy efficient than SMTJ-based bitstream generators that tune probabilities with spin currents and a factor of two more efficient than related CMOS-based implementations. The true randomness of this bitstream generator allows us to use them as the fundamental units of a novel neural network architecture. To take advantage of the potential savings, we codesign the algorithm with the circuit, rather than directly transcribing a classical neural network into hardware. The flexibility of the neural network mathematics allows us to adapt the network to the explicitly energy efficient choices we make at the device level. The result is a convolutional neural network design operating at $approx$ 150 nJ per inference with 97 % performance on MNIST -- a factor of 1.4 to 7.7 improvement in energy efficiency over comparable proposals in the recent literature.
rate research
Read More
Low-energy random number generation is critical for many emerging computing schemes proposed to complement or replace von Neumann architectures. However, current random number generators are always associated with an energy cost that is prohibitive for these computing schemes. In this paper, we introduce random number bit generation based on specific nanodevices: superparamagnetic tunnel junctions. We experimentally demonstrate high quality random bit generation that represents orders-of-magnitude improvements in energy efficiency compared to current solutions. We show that the random generation speed improves with nanodevice scaling, and investigate the impact of temperature, magnetic field and crosstalk. Finally, we show how alternative computing schemes can be implemented using superparamagentic tunnel junctions as random number generators. These results open the way for fabricating efficient hardware computing devices leveraging stochasticity, and highlight a novel use for emerging nanodevices.
The recently proposed probabilistic spin logic presents promising solutions to novel computing applications. Multiple cases of implementations, including invertible logic gate, have been studied numerically by simulations. Here we report an experimental demonstration of a magnetic tunnel junction-based hardware implementation of probabilistic spin logic.
Superparamagnetic tunnel junctions (SMTJs) are promising sources for the randomness required by some compact and energy-efficient computing schemes. Coupling SMTJs gives rise to collective behavior that could be useful for cognitive computing. We use a simple linear electrical circuit to mutually couple two SMTJs through their stochastic electrical transitions. When one SMTJ makes a thermally induced transition, the voltage across both SMTJs changes, modifying the transition rates of both. This coupling leads to significant correlation between the states of the two devices. Using fits to a generalized Neel-Brown model for the individual thermally bistable magnetic devices, we can accurately reproduce the behavior of the coupled devices with a Markov model.
Thermoelectric effects in magnetic nanostructures and the so-called spin caloritronics are attracting much interest. Indeed it provides a new way to control and manipulate spin currents which are key elements of spin-based electronics. Here we report on giant magnetothermoelectric effect in Al2O3 magnetic tunnel junctions. The thermovoltage in this geometry can reach 1 mV. Moreover a magneto-thermovoltage effect could be measured with ratio similar to the tunnel magnetoresistance ratio. The Seebeck coefficient can then be tuned by changing the relative magnetization orientation of the two magnetic layers in the tunnel junction. Therefore our experiments extend the range of spintronic devices application to thermoelectricity and provide a crucial piece of information for understanding the physics of thermal spin transport.
The feasibility of reservoir computing based on dipole-coupled nanomagnets is demonstrated using micro-magnetic simulations. The reservoir consists of an 2x10 array of nanomagnets. The static-magnetization directions of the nanomagnets are used as reservoir states. To update these states, we change the magnetization of one nanomagnet according to a single-bit-sequential signal. We also change the uniaxial anisotropy of the other nanomagnets using a voltage-induced magnetic-anisotropy change to enhance information flow, storage, and linear/nonlinear calculations. Binary tasks with AND, OR, and XOR operations were performed to evaluate the performance of the magnetic-array reservoir. The reservoir-computing output matrix was found to be trainable to perform AND, OR, and XOR operations with an input delay of up to three bits.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
