%0 Journal Article %J IEEE/ACM International Symposium on Microarchitecture (MICRO 2021) %D 2021 %T EdgeBERT: Sentence-Level Energy Optimizations for Latency-Aware Multi-Task NLP Inference %A Tambe, Thierry %A Coleman Hooper %A Lillian Pentecost %A Tianyu Jia %A Yang, En-Yu %A Marco Donato %A Victor Sanh %A Paul Whatmough %A Alexander M. Rush %A David Brooks %A Gu-Yeon Wei %X Transformer-based language models such as BERT provide significant accuracy improvement for a multitude of natural language processing (NLP) tasks. However, their hefty computational and memory demands make them challenging to deploy to resource-constrained edge platforms with strict latency requirements. We present EdgeBERT, an in-depth algorithm-hardware co-design for latency-aware energy optimization for multi-task NLP. EdgeBERT employs entropy-based early exit predication in order to perform dynamic voltage-frequency scaling (DVFS), at a sentence granularity, for minimal energy consumption while adhering to a prescribed target latency. Computation and memory footprint overheads are further alleviated by employing a calibrated combination of adaptive attention span, selective network pruning, and floating-point quantization. Furthermore, in order to maximize the synergistic benefits of these algorithms in always-on and intermediate edge computing settings, we specialize a 12nm scalable hardware accelerator system, integrating a fast-switching low-dropout voltage regulator (LDO), an all-digital phase-locked loop (ADPLL), as well as, high-density embedded non-volatile memories (eNVMs) wherein the sparse floating-point bit encodings of the shared multi-task parameters are carefully stored. Altogether, latency-aware multi-task NLP inference acceleration on the EdgeBERT hardware system generates up to 7x, 2.5x, and 53x lower energy compared to the conventional inference without early stopping, the latency-unbounded early exit approach, and CUDA adaptations on an Nvidia Jetson Tegra X2 mobile GPU, respectively. %B IEEE/ACM International Symposium on Microarchitecture (MICRO 2021) %G eng %U https://doi.org/10.48550/arXiv.2011.14203