Research map / Architectures and efficiency / RNNs and normalization
ASPEN: Breaking Operator Barriers for Efficient Parallelization of Deep Neural Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| increase tiles per layer from 1 to 128 (ablation) | ASPEN · synthetic 32-layer conv DNN | throughput (TFlops) | Helped |
| increase tiles per layer beyond 128 (ablation) | ASPEN · synthetic 32-layer conv DNN | throughput (TFlops) | Hurt |
| increase number of layers (ablation) | ASPEN · synthetic conv DNN, 4 to 256 layers | speedup | Helped |
| increase height, width and channel size (ablation) | ASPEN · synthetic conv DNN | speedup | No clear effect |
| increase batch size | ASPEN · CNN and NLP models, Threadripper 3990X | latency speedup | Helped |
| models with more multi-path layers (residual, multi-head attention) | ASPEN · Threadripper 3990X | latency speedup | Helped |
| use ASPEN instead of TorchScript | inference runtime · CNN and NLP models | latency speedup: up to 3.2× | Helped |
| use ASPEN on heterogeneous P and E cores instead of baselines | inference runtime · i9-12900K 1-16 threads | throughput scaling | Helped |
Closest papers by meaning
- DeepPCR: Parallelizing Sequential Operations in Neural Networks (NeurIPS 2023 · similarity 0.86)
- Auto-reconfiguration for Latency Minimization in CPU-based DNN Serving (ICML 2025 · similarity 0.85)
- Practical Performance Guarantees for Pipelined DNN Inference (ICML 2024 · similarity 0.85)
- Harnessing Manycore Processors with Distributed Memory for Accelerated Training of Sparse and Recurrent Models (AAAI 2024 · similarity 0.84)
- Synergistic Tensor and Pipeline Parallelism (NeurIPS 2025 · similarity 0.84)
- FlashRNN: I/O-Aware Optimization of Traditional RNNs on modern hardware (ICLR 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.