A post-training pipeline combining decomposition, pruning, and quantization compresses HDC models up to 100x with a 1-2% accuracy drop, plus an early-exit inference speedup.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DPQ-HD: Post-Training Compression for Ultra-Low Power Hyperdimensional Computing
A post-training pipeline combining decomposition, pruning, and quantization compresses HDC models up to 100x with a 1-2% accuracy drop, plus an early-exit inference speedup.