NetAdaptV2: Efficient Neural Architecture Search with Fast Super-Network Training and Architecture Optimization

• URL: http://arxiv.org/abs/2104.00031v1
• Date: Wed, 31 Mar 2021 18:03:46 GMT
• Title: NetAdaptV2: Efficient Neural Architecture Search with Fast Super-Network Training and Architecture Optimization
• Authors: Tien-Ju Yang, Yi-Lun Liao, Vivienne Sze
• Abstract summary: We present NetAdaptV2 with three innovations to better balance the time spent for each step while supporting non-differentiable search metrics. First, we propose channel-level bypass connections that merge network depth and layer width into a single search dimension. Second, ordered dropout is proposed to train multiple DNNs in a single forward-backward pass to decrease the time for training a super-network.
• Score: 15.63765190153914
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Neural architecture search (NAS) typically consists of three main steps: training a super-network, training and evaluating sampled deep neural networks (DNNs), and training the discovered DNN. Most of the existing efforts speed up some steps at the cost of a significant slowdown of other steps or sacrificing the support of non-differentiable search metrics. The unbalanced reduction in the time spent per step limits the total search time reduction, and the inability to support non-differentiable search metrics limits the performance of discovered DNNs. In this paper, we present NetAdaptV2 with three innovations to better balance the time spent for each step while supporting non-differentiable search metrics. First, we propose channel-level bypass connections that merge network depth and layer width into a single search dimension to reduce the time for training and evaluating sampled DNNs. Second, ordered dropout is proposed to train multiple DNNs in a single forward-backward pass to decrease the time for training a super-network. Third, we propose the multi-layer coordinate descent optimizer that considers the interplay of multiple layers in each iteration of optimization to improve the performance of discovered DNNs while supporting non-differentiable search metrics. With these innovations, NetAdaptV2 reduces the total search time by up to $5.8\times$ on ImageNet and $2.4\times$ on NYU Depth V2, respectively, and discovers DNNs with better accuracy-latency/accuracy-MAC trade-offs than state-of-the-art NAS works. Moreover, the discovered DNN outperforms NAS-discovered MobileNetV3 by 1.8% higher top-1 accuracy with the same latency. The project website is http://netadapt.mit.edu.

Related papers

• Optimizing DNN Inference on Multi-Accelerator SoCs at Training-time [5.05866540830123]
  We present ODiMO, a hardware-aware tool that efficiently explores fine-grain mapping of Deep Neural Networks (DNNs) among various on-chip CUs. We show that ODiMO reduces the latency of a DNN executed on the Darkside by up to 8x at iso-accuracy, compared to a manual mappings. When targeting energy, ODiMO produced up to 50.8x more efficient mappings, with minimal accuracy drop.
  arXiv  Detail & Related papers  (2024-09-27T09:10:44Z)
• Highly Efficient SNNs for High-speed Object Detection [7.3074002563489024]
  Experimental results show that our efficient SNN can achieve 118X speedup on GPU with only 1.5MB parameters for object detection tasks. We further verify our SNN on FPGA platform and the proposed model can achieve 800+FPS object detection with extremely low latency.
  arXiv  Detail & Related papers  (2023-09-27T10:31:12Z)
• Ultra-low Latency Adaptive Local Binary Spiking Neural Network with Accuracy Loss Estimator [4.554628904670269]
  We propose an ultra-low latency adaptive local binary spiking neural network (ALBSNN) with accuracy loss estimators. Experimental results show that this method can reduce storage space by more than 20 % without losing network accuracy.
  arXiv  Detail & Related papers  (2022-07-31T09:03:57Z)
• Towards Lossless ANN-SNN Conversion under Ultra-Low Latency with Dual-Phase Optimization [30.098268054714048]
  Spiking neural networks (SNNs) operating with asynchronous discrete events show higher energy efficiency with sparse computation. A popular approach for implementing deep SNNs is ANN-SNN conversion combining both efficient training of ANNs and efficient inference of SNNs. In this paper, we first identify that such performance degradation stems from the misrepresentation of the negative or overflow residual membrane potential in SNNs. Inspired by this, we decompose the conversion error into three parts: quantization error, clipping error, and residual membrane potential representation error.
  arXiv  Detail & Related papers  (2022-05-16T06:53:14Z)
• Sub-bit Neural Networks: Learning to Compress and Accelerate Binary Neural Networks [72.81092567651395]
  Sub-bit Neural Networks (SNNs) are a new type of binary quantization design tailored to compress and accelerate BNNs. SNNs are trained with a kernel-aware optimization framework, which exploits binary quantization in the fine-grained convolutional kernel space. Experiments on visual recognition benchmarks and the hardware deployment on FPGA validate the great potentials of SNNs.
  arXiv  Detail & Related papers  (2021-10-18T11:30:29Z)
• HANT: Hardware-Aware Network Transformation [82.54824188745887]
  We propose hardware-aware network transformation (HANT) HANT replaces inefficient operations with more efficient alternatives using a neural architecture search like approach. Our results on accelerating the EfficientNet family show that HANT can accelerate them by up to 3.6x with 0.4% drop in the top-1 accuracy on the ImageNet dataset.
  arXiv  Detail & Related papers  (2021-07-12T18:46:34Z)
• OPANAS: One-Shot Path Aggregation Network Architecture Search for Object Detection [82.04372532783931]
  Recently, neural architecture search (NAS) has been exploited to design feature pyramid networks (FPNs) We propose a novel One-Shot Path Aggregation Network Architecture Search (OPANAS) algorithm, which significantly improves both searching efficiency and detection accuracy.
  arXiv  Detail & Related papers  (2021-03-08T01:48:53Z)
• DNA: Differentiable Network-Accelerator Co-Search [36.68587348474986]
  We propose DNA, a Differentiable Network-Accelerator co-search framework for automatically searching for matched networks and accelerators. Specifically, DNA integrates two enablers: (1) a generic design space for DNN accelerators and compatible with DNN frameworks such as PyTorch to enable algorithmic exploration. Experiments and ablation studies show that the matched networks and accelerators generated by DNA consistently outperform state-of-the-art (SOTA) DNNs and accelerators.
  arXiv  Detail & Related papers  (2020-10-28T05:57:16Z)
• FATNN: Fast and Accurate Ternary Neural Networks [89.07796377047619]
  Ternary Neural Networks (TNNs) have received much attention due to being potentially orders of magnitude faster in inference, as well as more power efficient, than full-precision counterparts. In this work, we show that, under some mild constraints, computational complexity of the ternary inner product can be reduced by a factor of 2. We elaborately design an implementation-dependent ternary quantization algorithm to mitigate the performance gap.
  arXiv  Detail & Related papers  (2020-08-12T04:26:18Z)
• Neural Architecture Search For LF-MMI Trained Time Delay Neural Networks [61.76338096980383]
  A range of neural architecture search (NAS) techniques are used to automatically learn two types of hyper- parameters of state-of-the-art factored time delay neural networks (TDNNs) These include the DARTS method integrating architecture selection with lattice-free MMI (LF-MMI) TDNN training. Experiments conducted on a 300-hour Switchboard corpus suggest the auto-configured systems consistently outperform the baseline LF-MMI TDNN systems.
  arXiv  Detail & Related papers  (2020-07-17T08:32:11Z)
• FBNetV3: Joint Architecture-Recipe Search using Predictor Pretraining [65.39532971991778]
  We present an accuracy predictor that scores architecture and training recipes jointly, guiding both sample selection and ranking. We run fast evolutionary searches in just CPU minutes to generate architecture-recipe pairs for a variety of resource constraints. FBNetV3 makes up a family of state-of-the-art compact neural networks that outperform both automatically and manually-designed competitors.
  arXiv  Detail & Related papers  (2020-06-03T05:20:21Z)

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.