Memory device including 2-transistor memory cell structure for neural network

An apparatus comprising:

• a first memory cell including a first transistor and a second transistor, the first transistor including a first region and a first charge storage structure separated from the first region, the second transistor including a second region formed over the first charge storage structure, the first and second regions including materials of different conductivity types;
• a second memory cell including a first additional transistor and a second additional transistor, the first additional transistor including a first additional region and a second charge storage structure separated from the first additional region, the second additional transistor including a second additional region formed over the second charge storage structure, the first and second additional regions including materials of different conductivity types;
• and
• a memory control unit including circuitry configured to operate on the first and second memory cells to obtain multiply-accumulate information based on input information provided to access lines coupled to the first and second memory cells and weight information stored in the first and second memory cells.

A method comprising:

• obtaining first multiply-accumulate information from a first data line of a memory device, the first multiply-accumulate information based at least in part on first input information provided to a first access line coupled to a gate of a first transistor and a gate of second transistor of a first memory cell coupled to the first data line and based on first weight information associated with a neural network;
• obtaining second multiply-accumulate information from a second data line of the memory device, the second multiply-accumulate information based at least in part on second input information provided to a second access line coupled to a gate of a first transistor and a gate of second transistor of a second memory cell coupled to the second data line and based on second weight information associated with the neural network;
• and
• providing output information based on the first multiply-accumulate information and the second multiply-accumulate information.