用生成方式压缩神经网络权重，让微型设备跑得更快更省内存。
Once-for-All Channel Mixers (HYPERTINYPW): Generative Compression for TinyML
- 用微型MLP动态生成大部分权重，减少存储占用
- 在225KB内存下比1.4MB模型少84%字节，准确率仍超95%
- 适合资源受限的可穿戴设备和嵌入式传感任务
在微控制器上部署神经网络受限于千字节级的闪存和SRAM，即使经过INT8量化，1x1点卷积混合器仍占主导内存。我们提出HYPER-TINYPW，一种生成式压缩方法：将多数存储的点卷积权重替换为生成权重——一个共享的微型MLP在加载时一次性从极小的每层编码中合成点卷积核，缓存后用标准整数运算执行。该方法保持通用MCU运行时性能，仅增加一次合成开销；稳态延迟和能耗与INT8可分离CNN基线相当。通过跨层共享潜在基底消除冗余，同时保留INT8 PW1以稳定早期形态敏感混合。贡献包括：(i) 面向TinyML的打包字节计账，涵盖生成器、头/分解、编码、保留的PW1及主干；(ii) 统一评估框架，含验证调优的t*和置信区间；(iii) 可部署性分析，覆盖纯整数推理与启动/懒加载合成。在三个心电图基准（Apnea-ECG, PTB-XL, MIT-BIH）上，HYPER-TINYPW将宏F1与闪存的帕累托前沿向前推进：约225kB时匹配约1.4MB CNN，体积缩小6.31倍（减少84.15%字节），且保持至少95%的大模型宏F1。在32–64kB预算下仍能维持均衡检测，而紧凑基线模型性能下降。该机制广泛适用于其他1D生物信号、本地语音及嵌入式感知任务，其中每层冗余占主导，表明生成式压缩在资源受限系统中的广阔潜力。此外，该方法迁移至音频任务：在Speech Commands上达到96.2%测试准确率（最佳验证98.2%），支持其在嵌入式感知工作负载中的广泛应用。
原文摘要 · Abstract (English)
Deploying neural networks on microcontrollers is constrained by kilobytes of flash and SRAM, where 1x1 pointwise (PW) mixers often dominate memory even after INT8 quantization across vision, audio, and wearable sensing. We present HYPER-TINYPW, a compression-as-generation approach that replaces most stored PW weights with generated weights: a shared micro-MLP synthesizes PW kernels once at load time from tiny per-layer codes, caches them, and executes them with standard integer operators. This preserves commodity MCU runtimes and adds only a one-off synthesis cost; steady-state latency and energy match INT8 separable CNN baselines. Enforcing a shared latent basis across layers removes cross-layer redundancy, while keeping PW1 in INT8 stabilizes early, morphology-sensitive mixing. We contribute (i) TinyML-faithful packed-byte accounting covering generator, heads/factorization, codes, kept PW1, and backbone; (ii) a unified evaluation with validation-tuned t* and bootstrap confidence intervals; and (iii) a deployability analysis covering integer-only inference and boot versus lazy synthesis. On three ECG benchmarks (Apnea-ECG, PTB-XL, MIT-BIH), HYPER-TINYPW shifts the macro-F1 versus flash Pareto frontier: at about 225 kB it matches a roughly 1.4 MB CNN while being 6.31x smaller (84.15% fewer bytes), retaining at least 95% of large-model macro-F1. Under 32-64 kB budgets it sustains balanced detection where compact baselines degrade. The mechanism applies broadly to other 1D biosignals, on-device speech, and embedded sensing tasks where per-layer redundancy dominates, indicating a wider role for compression-as-generation in resource-constrained ML systems. Beyond ECG, HYPER-TINYPW transfers to TinyML audio: on Speech Commands it reaches 96.2% test accuracy (98.2% best validation), supporting broader applicability to embedded sensing workloads where repeated linear mixers dominate memory.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。