A Low-Voltage, Low-Power Reconfigurable Current-Mode Softmax Circuit for Analog Neural Networks

This paper presents a novel low-power low-voltage analog implementation of the softmax function, with electrically adjustable amplitude and slope parameters. We propose a modular design, which can be scaled by the number of inputs (and of corresponding outputs). It is composed of input current-voltage linear converter stages (1st stages), MOSFETs operating in a subthreshold regime implementing the exponential functions (2nd stages), and analog divider stages (3rd stages). Each stage is only composed of p-type MOSFET transistors. Designed in a 0.18 mu m CMOS technology (TSMC), the proposed softmax circuit can be operated at a supply voltage of 500 mV. A ten-input/ten-output realization occupies a chip area of 2570 mu m(2) and consumes only 3 mu W of power, representing a very compact and energy-efficient option compared to the corresponding digital implementations.