SOT-based spin torque oscillators for oscillatory neural networks
An apparatus is provided that includes an array including m rows and n columns of nodes. Each column of nodes is coupled to one of n first conductive lines, and each row of nodes is coupled to one of m second conductive lines. Each node of the m rows and n columns of nodes includes a spin orbit torque-based spin torque oscillator circuit configured to oscillate at a corresponding intrinsic frequency. The spin orbit torque-based spin torque oscillator circuits are configured to generate m output signals at the m second conductive lines upon application of n input signals to corresponding n first conductive lines. The n input signals correspond to an n-element input vector, and each input signal includes a corresponding input signal frequency. Each of the m output signals include frequency domain components at the input signal frequencies. The magnitudes of the frequency domain components at the input signal frequencies depend on a degree of synchronization between the input signal frequencies and the intrinsic frequencies.
1. An apparatus comprising:
an array comprising m rows and n columns of nodes, each column of nodes coupled to one of n first conductive lines, each row of nodes coupled to one of m second conductive lines, each node of the m rows and n columns of nodes comprising a spin orbit torque-based spin torque oscillator circuit configured to oscillate at a corresponding intrinsic frequency,
wherein:
the spin orbit torque-based spin torque oscillator circuits are configured to generate m output signals at the m second conductive lines upon application of n input signals to corresponding n first conductive lines, the n input signals corresponding to an n-element input vector, each input signal comprising a corresponding input signal frequency; and
each of the m output signals comprises frequency domain components at the input signal frequencies, and magnitudes of the frequency domain components at the input signal frequencies depend on a degree of synchronization between the input signal frequencies and the intrinsic frequencies.
2. The apparatus of claim 1 , wherein the corresponding intrinsic frequency of each spin orbit torque-based spin torque oscillator circuit is programmable.
3. The apparatus of claim 1 , wherein each spin orbit torque-based spin torque oscillator circuit is configured to oscillate at corresponding oscillation frequencies based on signals input to the spin orbit torque-based spin torque oscillator circuits.
4. The apparatus of claim 1 , wherein the corresponding intrinsic frequency of each spin orbit torque-based spin torque oscillator circuit is based on values of dc input currents applied to the spin orbit torque-based spin torque oscillator circuits.
5. The apparatus of claim 1 , wherein the spin orbit torque-based spin torque oscillator circuits are programmed to oscillate at the corresponding intrinsic frequencies based on corresponding weights of an m×n array of weights.
6. The apparatus of claim 5 , wherein each of the weights represents an amplitude of dc currents injected into the spin orbit torque-based spin torque oscillator circuits to program the spin orbit torque-based spin torque oscillator circuits to oscillate at the corresponding intrinsic frequencies.
7. The apparatus of claim 1 , wherein each spin orbit torque-based spin torque oscillator circuit is coupled to receive one of the n input signals comprising a corresponding input signal frequency, and is configured to oscillate at the corresponding input signal frequency if a difference between the corresponding input signal frequency and the corresponding intrinsic frequency is less than or equal to a predetermined threshold.
8. The apparatus of claim 1 , wherein each spin orbit torque-based spin torque oscillator circuit is coupled to receive one of the n input signals comprising a corresponding input signal frequency, and is configured to synchronize its oscillation frequency to the corresponding input signal frequency.
9. The apparatus of claim 1 , wherein each spin orbit torque-based spin torque oscillator circuit comprises a spin Hall effect layer comprising one or more of a heavy metal and a topological insulator.
10. The apparatus of claim 1 , wherein each spin orbit torque-based spin torque oscillator circuit comprises:
a spin Hall effect layer; and
a magnetic tunnel junction comprising a pinned layer and a free layer that each have a perpendicular magnetization direction, wherein the magnetic tunnel junction comprises a substantially cylindrical shape.
11. The apparatus of claim 1 , wherein each spin orbit torque-based spin torque oscillator circuit comprises:
a spin Hall effect layer; and
a magnetic tunnel junction comprising a pinned layer and a free layer that each have an in-plane magnetization direction,
wherein the magnetic tunnel junction comprises a substantially ellipsoidal shape, and the pinned layer and the free layer each have an easy axis perpendicular to a current flow in the spin Hall effect layer.
12. The apparatus of claim 1 , wherein each spin orbit torque-based spin torque oscillator circuit comprises:
a spin Hall effect layer; and
a magnetic tunnel junction comprising a pinned layer and a free layer that each have an in-plane magnetization direction,
wherein the magnetic tunnel junction comprises a substantially ellipsoidal shape, and the pinned layer and the free layer each have an easy axis that is at an angle off-perpendicular to a current flow in the spin Hall effect layer.
13. An apparatus comprising:
a cross-point array comprising a plurality of spin orbit torque-based spin torque oscillator circuits configured to store synaptic weights of an artificial neural network;
a plurality of first conductive lines coupled to the spin orbit torque-based spin torque oscillator circuits; and
a plurality of second conductive lines coupled to the spin orbit torque-based spin torque oscillator circuits,
wherein in response to a plurality of input currents coupled to the plurality of first conductive lines, the spin orbit torque-based spin torque oscillator circuits generate output currents at the plurality of second conductive lines representing outputs of the artificial neural network, and
wherein each output current comprises frequency domain components each comprising a magnitude that depends on how closely frequencies of the input currents match oscillation frequencies of the spin orbit torque-based spin torque oscillator circuits.
14. The apparatus of claim 13 , wherein each spin orbit torque-based spin torque oscillator circuit comprises a first terminal coupled to one of the first conductive lines, and a second terminal coupled to one of the second conductive lines.
15. An apparatus comprising:
a cross-point array comprising a plurality of spin orbit torque-based spin torque oscillator circuits configured to store synaptic weights of an artificial neural network;
a plurality of first conductive lines coupled to the spin orbit torque-based spin torque oscillator circuits; and
a plurality of second conductive lines coupled to the spin orbit torque-based spin torque oscillator circuits,
wherein in response to a plurality of input currents coupled to the plurality of first conductive lines, the spin orbit torque-based spin torque oscillator circuits generate output currents at the plurality of second conductive lines representing outputs of the artificial neural network; and
wherein each spin orbit torque-based spin torque oscillator circuit comprises a spin Hall effect layer and a magnetic tunnel junction comprising a free layer, wherein the free layer comprises a magnetization direction that oscillates at a programmable frequency.
16. An apparatus comprising:
a cross-point array comprising a plurality of spin orbit torque-based spin torque oscillator circuits configured to store synaptic weights of an artificial neural network;
a plurality of first conductive lines coupled to the spin orbit torque-based spin torque oscillator circuits; and
a plurality of second conductive lines coupled to the spin orbit torque-based spin torque oscillator circuits,
wherein in response to a plurality of input currents coupled to the plurality of first conductive lines, the spin orbit torque-based spin torque oscillator circuits generate output currents at the plurality of second conductive lines representing outputs of the artificial neural network; and
wherein each spin orbit torque-based spin torque oscillator circuit is programmed to oscillate at a corresponding intrinsic frequency based on corresponding weights of an m×n array of weights of the artificial neural network.
17. The apparatus of claim 13 , wherein each spin orbit torque-based spin torque oscillator circuit comprises a programmable oscillation frequency.
18. An apparatus comprising:
a cross-point array comprising a plurality of spin orbit torque-based spin torque oscillator circuits configured to store synaptic weights of an artificial neural network;
a plurality of first conductive lines coupled to the spin orbit torque-based spin torque oscillator circuits; and
a plurality of second conductive lines coupled to the spin orbit torque-based spin torque oscillator circuits,
wherein in response to a plurality of input currents coupled to the plurality of first conductive lines, the spin orbit torque-based spin torque oscillator circuits generate output currents at the plurality of second conductive lines representing outputs of the artificial neural network; and
wherein each spin orbit torque-based spin torque oscillator circuit further comprises a third terminal coupled to GROUND.
19. A method comprising:
programming each of a first plurality of spin orbit torque-based spin torque oscillator circuits to oscillate at a corresponding intrinsic frequency;
generating n input currents corresponding to an n-element input vector, each input current comprising a corresponding input signal frequency; and
coupling the n input currents to the first plurality of spin orbit torque-based spin torque oscillator circuits to generate m output currents, each output current comprising frequency domain components at the input signal frequencies, and magnitudes of the frequency domain components at the input signal frequencies depend on a degree of synchronization between the input signal frequencies and the intrinsic frequencies.
20. The method of claim 19 , further comprising providing them output currents as input currents to a second plurality of spin orbit torque-based spin torque oscillator circuits that are each programmed to oscillate at a corresponding intrinsic frequency.