Multibit neural inference in a N-ary crossbar architecture
For hardware designers, this work provides a simulation framework and analysis of error sources in N-ary crossbar architectures for neural network inference.
This work presents a simulation framework for N-ary crossbar architectures using magnetic tunnel junctions for in-memory computing. The system achieves 94.48% accuracy on MNIST (vs. 97.56% software baseline), with weight quantization identified as the primary error source.
In-memory computing (IMC) enables energy-efficient neural network inference by computing analog matrix-vector multiplications (MVM) in memory crossbar arrays. In this work we present a simulation framework for N-ary crossbar architectures that retrieves MVM results with minimal implementation assumptions. The XOR and MNIST classification tasks were successfully inferred using a simulated crossbar array of (4x4) 4-states magnetic tunnel junctions (MTJ). MNIST accuracy reached 94.48% (vs. 97.56% software baseline). The software-hardware performance gap was further reduced using PCA dimensionality reduction. We identified weight quantization as the primary error source, and studied its impact alongside systematic nonidealities and random noise. We find that cell-specific random noise is less detrimental than systematic errors due to averaging across the array. Finally, we demonstrate an optimal number of states per cell that balances quantization error against resistance state resolution to minimize total MVM error.