DCP: Learning Accelerator Dataflow for Neural Network via Propagation

• URL: http://arxiv.org/abs/2410.06553v1
• Date: Wed, 9 Oct 2024 05:16:44 GMT
• Title: DCP: Learning Accelerator Dataflow for Neural Network via Propagation
• Authors: Peng Xu, Wenqi Shao, Mingyu Ding, Ping Luo,
• Abstract summary: This work proposes an efficient data-centric approach, named Dataflow Code Propagation (DCP), to automatically find the optimal dataflow for DNN layers in seconds without human effort. DCP learns a neural predictor to efficiently update the dataflow codes towards the desired gradient directions to minimize various optimization objectives. For example, without using additional training data, DCP surpasses the GAMMA method that performs a full search using thousands of samples.
• Score: 52.06154296196845
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Deep neural network (DNN) hardware (HW) accelerators have achieved great success in improving DNNs' performance and efficiency. One key reason is dataflow in executing a DNN layer, including on-chip data partitioning, computation parallelism, and scheduling policy, which have large impacts on latency and energy consumption. Unlike prior works that required considerable efforts from HW engineers to design suitable dataflows for different DNNs, this work proposes an efficient data-centric approach, named Dataflow Code Propagation (DCP), to automatically find the optimal dataflow for DNN layers in seconds without human effort. It has several attractive benefits that prior arts do not have. (i) We translate the HW dataflow configuration into a code representation in a unified dataflow coding space, which can be optimized by backpropagating gradients given a DNN layer or network. (ii) DCP learns a neural predictor to efficiently update the dataflow codes towards the desired gradient directions to minimize various optimization objectives e.g., latency and energy. (iii) It can be easily generalized to unseen HW configurations in a zero-shot or few-shot learning manner. For example, without using additional training data, DCP surpasses the GAMMA method that performs a full search using thousands of samples. Extensive experiments on several representative models such as MobileNet, ResNet, and ViT show that DCP outperforms its counterparts in various settings.

Related papers

• HASS: Hardware-Aware Sparsity Search for Dataflow DNN Accelerator [47.66463010685586]
  We propose a novel approach to exploit unstructured weights and activations sparsity for dataflow accelerators, using software and hardware co-optimization. We achieve an efficiency improvement ranging from 1.3$times$ to 4.2$times$ compared to existing sparse designs.
  arXiv  Detail & Related papers  (2024-06-05T09:25:18Z)
• Algorithm-Hardware Co-Design of Distribution-Aware Logarithmic-Posit Encodings for Efficient DNN Inference [4.093167352780157]
  We introduce Logarithmic Posits (LP), an adaptive, hardware-friendly data type inspired by posits. We also develop a novel genetic-algorithm based framework, LP Quantization (LPQ), to find optimal layer-wise LP parameters.
  arXiv  Detail & Related papers  (2024-03-08T17:28:49Z)
• A Multi-Head Ensemble Multi-Task Learning Approach for Dynamical Computation Offloading [62.34538208323411]
  We propose a multi-head ensemble multi-task learning (MEMTL) approach with a shared backbone and multiple prediction heads (PHs) MEMTL outperforms benchmark methods in both the inference accuracy and mean square error without requiring additional training data.
  arXiv  Detail & Related papers  (2023-09-02T11:01:16Z)
• KAPLA: Pragmatic Representation and Fast Solving of Scalable NN Accelerator Dataflow [0.0]
  We build a generic, optimized, and fast dataflow solver, KAPLA, to explore the design space with effective validity check and efficiency estimation. KAPLA achieves within only 2.2% and 7.7% energy overheads on the result dataflow for training and inference. It also outperforms random and machine-learning-based approaches, with more optimized results and orders of magnitude faster search speedup.
  arXiv  Detail & Related papers  (2023-06-09T03:12:42Z)
• Intelligence Processing Units Accelerate Neuromorphic Learning [52.952192990802345]
  Spiking neural networks (SNNs) have achieved orders of magnitude improvement in terms of energy consumption and latency. We present an IPU-optimized release of our custom SNN Python package, snnTorch.
  arXiv  Detail & Related papers  (2022-11-19T15:44:08Z)
• Two-Timescale End-to-End Learning for Channel Acquisition and Hybrid Precoding [94.40747235081466]
  We propose an end-to-end deep learning-based joint transceiver design algorithm for millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) systems. We develop a DNN architecture that maps the received pilots into feedback bits at the receiver, and then further maps the feedback bits into the hybrid precoder at the transmitter.
  arXiv  Detail & Related papers  (2021-10-22T20:49:02Z)
• JUMBO: Scalable Multi-task Bayesian Optimization using Offline Data [86.8949732640035]
  We propose JUMBO, an MBO algorithm that sidesteps limitations by querying additional data. We show that it achieves no-regret under conditions analogous to GP-UCB. Empirically, we demonstrate significant performance improvements over existing approaches on two real-world optimization problems.
  arXiv  Detail & Related papers  (2021-06-02T05:03:38Z)
• ConfuciuX: Autonomous Hardware Resource Assignment for DNN Accelerators using Reinforcement Learning [5.251940442946459]
  We propose an autonomous strategy called ConfuciuX to find optimized HW resource assignments for a given model and dataflow style. It converges to the optimized hardware configuration 4.7 to 24 times faster than alternate techniques.
  arXiv  Detail & Related papers  (2020-09-04T04:59:26Z)
• DRACO: Co-Optimizing Hardware Utilization, and Performance of DNNs on Systolic Accelerator [5.65116500037191]
  We propose data reuse computation aware co-optimization (DRACO) DRACO improves the PE utilization of memory-bound DNNs without any additional need for dataflow/micro-architecture modifications. Unlike the previous co-optimization methods, DRACO not only maximizes performance and energy efficiency but also improves the predictive performance of DNNs.
  arXiv  Detail & Related papers  (2020-06-26T17:06:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.