mlsanityData to 7 Oct 2026

Pruning In Time (PIT): A Lightweight Network Architecture Optimizer for Temporal Convolutional Networks

arXiv 2203.14768 ↗First posted 28 Mar 2022cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

Temporal Convolutional Networks (TCNs) are promising Deep Learning models for time-series processing tasks. One key feature of TCNs is time-dilated convolution, whose optimization requires extensive experimentation. We propose an automatic dilation optimizer, which tackles the problem as a weight pruning on the time-axis, and learns dilation factors together with weights, in a single training. Our method reduces the model size and inference latency on a real SoC hardware target by up to 7.4x and 3x, respectively with no accuracy drop compared to a network without dilation. It also yields a rich set of Pareto-optimal TCNs starting from a single model, outperforming hand-designed solutions in both size and accuracy.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 12 of its 20 references that are in the map. First posted on arXiv 28 Mar 2022.

Builds on

12 of 20 references.

References · 12

  1. 2018DARTS: Differentiable Architecture Searchcited by 1,782
  2. 2016Neural Architecture Search with Reinforcement Learningcited by 1,489
  3. 2018An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modelingcited by 1,447
  4. 2018MnasNet: Platform-Aware Neural Architecture Search for Mobilecited by 937
  5. 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 778
  6. 2018ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardwarecited by 666
  7. 2020MCUNet: Tiny Deep Learning on IoT Devicescited by 269
  8. 2015A Critical Review of Recurrent Neural Networks for Sequence Learningcited by 223
  9. 2020FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel Dimensionscited by 152
  10. 2016Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1cited by 102
  11. 2017MorphNet: Fast & Simple Resource-Constrained Structure Learning of Deep Networkscited by 92
  12. 2017Deep Neural Networks for Multiple Speaker Detection and Localizationcited by 21

Built on by

Cited by year

Citing papers per year: 2022: 3; 2023: 1.

4 papers.

  1. 2023Lightweight Neural Architecture Search for Temporal Convolutional Networks at the Edgecited by 15
  2. 2022Human Activity Recognition on Microcontrollers with Quantized and Adaptive Deep Neural Networkscited by 6
  3. 2022Multi-Complexity-Loss DNAS for Energy-Efficient and Memory-Constrained Deep Neural Networkscited by 2
  4. 2022Privacy-preserving Social Distance Monitoring on Microcontrollers with Low-Resolution Infrared Sensors and CNNscited by 1