Error Correctable Range-Addressable Lookup for Activation and Quantization in AI Automotive Electronics

In this paper, we generalize the lightweight-slope piecewise-line range-addressable lookup table for efficient approximate computing in any activation/quantization function, and propose an error-correcting algorithm and circuitry using the AN codes for enhancing reliability. A BLER/SNR simulation proves the SEC/DEC capability of the firstly-presented AN-coded neuron. Comparisons with similar state-of-the-art works show that the proposed technique is the most efficient and error-correctable lookup table for any function in a medium resolution within 8-12 bits.