A memory circuit configured to perform multiply-accumulate (MAC) operations for performance of an artificial neural network includes a series of synapse cells arranged in a cross-bar array. Each cell includes a memory transistor connected in series with a memristor. The memory circuit also includes input lines connected to the source terminal of the memory transistor in each cell, output lines connected to an output terminal of the memristor in each cell, and programming lines coupled to a gate terminal of the memory transistor in each cell. The memristor of each cell is configured to store a conductance value representative of a synaptic weight of a synapse connected to a neuron in the artificial neural network, and the memory transistor of each cell is configured to store a threshold voltage representative of a synaptic importance value of the synapse connected to the neuron in the artificial neural network.

Numerous embodiments are disclosed of a non-volatile memory cell array formed in a p-well, which is formed in a deep n-well, which is formed in a p-substrate. During an erase operation, a negative voltage is applied to the p-well, which reduces the peak positive voltage required to be applied to the cells to cause the cells to erase.

Numerous embodiments are disclosed for splitting an array of non-volatile memory cells in an analog neural memory in a deep learning artificial neural network into multiple parts. Each part of the array interacts with certain circuitry dedicated to that part and with other circuitry that is shared with one or more other parts of the array.

Numerous embodiments of a precision tuning algorithm and apparatus are disclosed for precisely and quickly depositing the correct amount of charge on the floating gate of a non-volatile memory cell within a vector-by-matrix multiplication (VMM) array in an artificial neural network. Selected cells thereby can be programmed with extreme precision to hold one of N different values.

The present invention relates to a flash memory device that uses strap cells in a memory array of non-volatile memory cells as source line pull down circuits. In one embodiment, the strap cells are erase gate strap cells. In another embodiment, the strap cells are source line strap cells. In another embodiment, the strap cells are control gate strap cells. In another embodiment, the strap cells are word line strap cells.

A non-volatile memory device is provided. The non-volatile memory device includes a substrate, a floating gate, and a gate. The substrate includes a source region and a drain region, and a channel region between the source region and the drain region. The floating gate is over the channel region. The floating gate includes a first conductive layer and a second conductive layer underlying the first conductive layer. The gate is adjacent to the floating gate.

Various embodiments of high voltage generation circuits, high voltage operational amplifiers, adaptive high voltage supplies, adjustable high voltage incrementor, adjustable reference supplies, and reference circuits are disclosed. These circuits optionally can be used for programming a non-volatile memory cell in an analog neural memory to store one of many possible values.

A stacked-gate non-volatile memory cell includes a semiconductor substrate, a floating gate, a first spacer, a control gate, a second spacer, a first doped region and a second doped region. The floating gate is formed over the semiconductor substrate. The first spacer is contacted with a sidewall of the floating gate. The control gate is formed on a top side and a lateral side of the floating gate. The control gate is not contacted with the floating gate. The second spacer is contacted with a sidewall of the control gate. The first doped region and the second doped region are formed in the surface of the semiconductor substrate, and respectively located at two sides of the floating gate.

A memory circuit configured to perform multiply-accumulate (MAC) operations for performance of an artificial neural network includes a series of synapse cells arranged in a cross-bar array. Each cell includes a memory transistor connected in series with a memristor. The memory circuit also includes input lines connected to the source terminal of the memory transistor in each cell, output lines connected to an output terminal of the memristor in each cell, and programming lines coupled to a gate terminal of the memory transistor in each cell. The memristor of each cell is configured to store a conductance value representative of a synaptic weight of a synapse connected to a neuron in the artificial neural network, and the memory transistor of each cell is configured to store a threshold voltage representative of a synaptic importance value of the synapse connected to the neuron in the artificial neural network.

Numerous examples of an input function circuit block and an output neuron circuit block coupled to a vector-by-matrix multiplication (VMM) array in an artificial neural network are disclosed. In one example, an artificial neural network comprises a vector-by-matrix multiplication array comprising a plurality of non-volatile memory cells organized into rows and columns; an input function circuit block to receive digital input signals, convert the digital input signals into analog signals, and apply the analog signals to control gate terminals of non-volatile memory cells in one or more rows of the array during a programming operation; and an output neuron circuit block to receive analog currents from the columns of the array during a read operation and generate an output signal.