Programmable output blocks for analog neural memory in a deep learning artificial neural network
AssigneeSilicon Storage Technology, Inc.
InventorHieu Van Tran
Life of this patent
Estimating the term…
Abstract
Numerous embodiments are disclosed for programmable output blocks for use with a VMM array within an artificial neural network. In one embodiment, the gain of an output block can be configured by a configuration signal. In another embodiment, the resolution of an ADC in the output block can be configured by a configuration signal.
Claims12
A programmable neuron output block for generating an output of a neural network memory array, comprising:
- one or more input nodes for receiving current from a neural network memory array;
- and
- a gain configuration circuit for receiving a gain configuration signal and applying a gain factor to the received current in response to the gain configuration signal to generate an output, wherein the gain configuration signal comprises an analog signal.
A programmable neuron output block for generating an output of a neural network memory array, comprising:
- one or more input nodes for receiving current from a neural network memory array;
- and
- a gain configuration circuit for receiving a gain configuration signal and applying a gain factor to the received current in response to the gain configuration signal to generate an output, wherein the gain configuration circuit comprises a variable resistor controlled by the gain configuration signal.
A programmable neuron output block for generating an output of a neural network memory array, comprising:
- one or more input nodes for receiving current from a neural network memory array;
- and
- a gain configuration circuit for receiving a gain configuration signal and applying a gain factor to the received current in response to the gain configuration signal to generate an output, wherein the gain configuration circuit comprises a variable capacitor controlled by the gain configuration signal.
A programmable neuron output block for generating an output of a neural network memory array, comprising:
- one or more input nodes for receiving current from a neural network memory array;
- and
- a gain configuration circuit for receiving a gain configuration signal and applying a gain factor to the received current in response to the gain configuration signal to generate an output, wherein the gain configuration signal depends on the number or rows or columns enabled in the neural network memory array from which the current was received.
A programmable neuron output block for generating an output of a neural network memory array, comprising:
- an analog-to-digital converter for receiving current from the neural network memory array and a control signal and generating a digital output, wherein resolution of the digital output is determined by the control signal.
A programmable neuron output block for generating an output of a neural network memory array, comprising:
- a hybrid analog-to-digital converter for converting the output of the neural network memory array into a digital output, the hybrid analog-to-digital converter comprising:
- a first analog-to-digital converter for generating a first portion of the digital output;
- and
- a second analog-to-digital converter for generating a second portion of the digital output.
Dependencies are inferred from the claim text.