Research map / Architectures and efficiency / Quantization
Quantized Memory-Augmented Neural Networks
Seongsik Park, Seijoon Kim, Seil Lee, Ho Bae, Sungroh Yoon
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Replace cosine similarity with bounded Hamming similarity for memory addressing | MANN with 8-bit fixed-point quantization · bAbI 10k | average test error rate (mean): 51.23 -> 30.43 | Improved |
| Add early stopping (ablation) | MANN with 8-bit fixed-point quantization · bAbI 10k | average test error rate (mean): 51.23 -> 43.87 | Improved |
| Add memory controller quantization control after early stopping (ablation) | MANN with 8-bit fixed-point quantization · bAbI 10k | average test error rate (mean): 43.87 -> 38.50 | Improved |
| Add early stopping (ablation) | Q-MANN with 8-bit fixed-point quantization · bAbI 10k | average test error rate (mean): 30.43 -> 32.88 | Hurt |
| Add memory controller quantization control with early stopping (ablation) | Q-MANN with 8-bit fixed-point quantization · bAbI 10k | average test error rate (mean): 32.88 -> 27.68 | Improved |
| Use binary activations instead of 8-bit fixed-point activations (ablation) | MANN with 8-bit fixed-point parameters · bAbI 10k | average test error rate (mean): 51.23 -> 44.49 | Improved |
| Add early stopping (ablation) | MANN with binary activations and 8-bit fixed-point parameters · bAbI 10k | average test error rate (mean): 44.49 -> 37.12 | Improved |
| Add memory controller quantization control after early stopping (ablation) | MANN with binary activations and 8-bit fixed-point parameters · bAbI 10k | average test error rate (mean): 37.12 -> 42.24 | Hurt |
| Replace cosine similarity with Hamming similarity and add early stopping plus memory controller quantization control (ablation) | MANN with binary activations and 8-bit fixed-point parameters · bAbI 10k | average test error rate (mean): 42.24 -> 31.00 | Improved |
Closest papers by meaning
- Alternating Multi-bit Quantization for Recurrent Neural Networks (ICLR 2018 · similarity 0.86)
- And the Bit Goes Down: Revisiting the Quantization of Neural Networks (ICLR 2020 · similarity 0.85)
- Adaptive Quantization for Deep Neural Network (AAAI 2018 · similarity 0.85)
- LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networks (ECCV 2018 · similarity 0.85)
- Quantization without Tears (CVPR 2025 · similarity 0.85)
- ParetoQ: Improving Scaling Laws in Extremely Low-bit LLM Quantization (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
