Abstract
Test-Time Augmentation (TTA) is a popular technique that aims to improve the accuracy of Convolutional Neural Networks (ConvNets) at inference-time. TTA addresses a limitation inherent to any deep learning pipeline, that is, training datasets cover only a tiny portion of the possible inputs. For this reason, when ported to real-life scenarios, ConvNets may suffer from substantial accuracy loss due to unseen input patterns received under unpredictable external conditions that can mislead the model. TTA tackles this problem directly on the field, first running multiple inferences on a set of altered versions of the same input sample and then computing the final outcome through a consensus of the aggregated predictions. TTA has been conceived to run on cloud systems powered with high-performance GPUs, where the altered inputs get processed in parallel with no (or negligible) performance overhead. Unfortunately, when shifted on embedded CPUs, TTA introduces latency penalties that limit its adoption for edge applications. For a more efficient resource usage, we can rely on an adaptive implementation of TTA, AdapTTA, that adjusts the number of inferences dynamically, depending on the input complexity. In this work, we assess the figures of merit of the AdapTTA framework, exploring different configurations of its basic blocks, i.e., the augmentation policy, the predictions aggregation function, and the model confidence score estimator, suitable for the integration with the proposed adaptive system. We conducted an extensive experimental evaluation, considering state-of-the-art ConvNets for image classification, MobileNets and EfficientNets, deployed onto a commercial embedded device, the ARM Cortex-A CPU. The collected results reveal that thanks to optimal design choices, AdapTTA ensures substantial acceleration compared to a static TTA, with up to 2.21\(\times \) faster processing preserving the same accuracy level. This comprehensive analysis helps designers identify the most efficient AdapTTA configuration for custom inference engines running on the edge.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Hendrycks, D., Dietterich, T.G.: Benchmarking neural network robustness to common corruptions and perturbations. In: 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, 6–9 May 2019 (2019)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems 25: 26th Annual Conference on Neural Information Processing Systems 2012. Proceedings of a Meeting Held, Lake Tahoe, Nevada, USA, 3–6 December 2012, pp. 1106–1114 (2012)
Simard, P.Y., Steinkraus, D., Platt, J.C.: Best practices for convolutional neural networks applied to visual document analysis. In: 7th International Conference on Document Analysis and Recognition (ICDAR 2003), Edinburgh, Scotland, UK, 3–6 August 2003, vol. 2, pp. 958–962. IEEE Computer Society (2003)
Cubuk, E.D., Zoph, B., Shlens, J., Le, Q.V.: RandAugment: practical automated data augmentation with a reduced search space. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR Workshops 2020, Seattle, WA, USA, 14–19 June 2020, pp. 3008–3017. Computer Vision Foundation/IEEE (2020)
Cubuk, E.D., Zoph, B., Mané, D., Vasudevan, V., Le, Q.V.: AutoAugment: learning augmentation strategies from data. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2019, Long Beach, CA, USA, 16–20 June 2019, pp. 113–123. Computer Vision Foundation/IEEE (2019)
Hataya, R., Zdenek, J., Yoshizoe, K., Nakayama, H.: Faster AutoAugment: learning augmentation strategies using backpropagation. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12370, pp. 1–16. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58595-2_1
Tan, M., Le, Q.V.: EfficientNet: rethinking model scaling for convolutional neural networks. In: Proceedings of the 36th International Conference on Machine Learning, ICML 2019, Long Beach, California, USA, 9–15 June 2019. Proceedings of Machine Learning Research, vol. 97, pp. 6105–6114. PMLR (2019)
Lyzhov, A., Molchanova, Y., Ashukha, A., Molchanov, D., Vetrov, D.P.: Greedy policy search: a simple baseline for learnable test-time augmentation. In: Proceedings of the Thirty-Sixth Conference on Uncertainty in Artificial Intelligence, UAI 2020, Virtual, 3–6 August 2020. Proceedings of Machine Learning Research, vol. 124, pp. 1308–1317. AUAI Press (2020)
Recht, B., Roelofs, R., Schmidt, L., Shankar, V.: Do imagenet classifiers generalize to imagenet? In: Proceedings of the 36th International Conference on Machine Learning, ICML 2019, Long Beach, California, USA, 9–15 June 2019. Proceedings of Machine Learning Research, vol. 97, pp. 5389–5400. PMLR (2019)
Sun, Y., et al.: Test-time training with self-supervision for generalization under distribution shifts. In: Proceedings of the 37th International Conference on Machine Learning, ICML 2020, Virtual, 13–18 July 2020. Proceedings of Machine Learning Research, vol. 119, pp. 9229–9248. PMLR (2020)
Kim, J., Kim, H., Kim, G.: Model-agnostic boundary-adversarial sampling for test-time generalization in few-shot learning. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12346, pp. 599–617. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58452-8_35
Howard, A.G.: Some improvements on deep convolutional neural network based image classification. In: 2nd International Conference on Learning Representations, ICLR 2014, Banff, AB, Canada, 14–16 April 2014. Conference Track Proceedings (2014)
Szegedy, C., et al.: Going deeper with convolutions. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015, Boston, MA, USA, 7–12 June 2015, pp. 1–9. IEEE Computer Society (2015)
Peluso, V., Rizzo, R.G., Cipolletta, A., Calimera, A.: Inference on the edge: performance analysis of an image classification task using off-the-shelf CPUs and open-source convnets. In: Sixth International Conference on Social Networks Analysis, Management and Security, SNAMS 2019, Granada, Spain, 22–25 October 2019, pp. 454–459. IEEE (2019)
Peluso, V., Rizzo, R.G., Calimera, A.: Performance profiling of embedded convnets under thermal-aware DVFs. Electronics 8(12), 1423 (2019)
Grimaldi, M., Peluso, V., Calimera, A.: Optimality assessment of memory-bounded convnets deployed on resource-constrained RISC cores. IEEE Access 7, 152 599–152 611 (2019)
Mocerino, L., Rizzo, R.G., Peluso, V., Calimera, A., Macii, E.: AdapTTA: adaptive test-time augmentation for reliable embedded convnets. In: 29th IFIP/IEEE International Conference on Very Large Scale Integration, VLSI-SoC 2021, Singapore, Singapore, 4–7 October 2021, pp. 1–6. IEEE (2021)
Devries, T., Taylor, G.W.: Improved regularization of convolutional neural networks with cutout. CoRR, vol. abs/1708.04552 (2017)
Zhang, H., Cissé, M., Dauphin, Y.N., Lopez-Paz, D.: Mixup: beyond empirical risk minimization. In: 6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, 30 April–3 May 2018. Conference Track Proceedings (2018)
Yun, S., et al.: CutMix: regularization strategy to train strong classifiers with localizable features. In: 2019 IEEE/CVF International Conference on Computer Vision, ICCV 2019, Seoul, Korea (South), 27 October–2 November 2019, pp. 6022–6031. IEEE (2019)
Tan, M., Le, Q.V.: EfficientNetV2: smaller models and faster training. In: Proceedings of the 38th International Conference on Machine Learning, ICML 2021, Virtual, 18–24 July 2021. Proceedings of Machine Learning Research, vol. 139, pp. 10 096–10 106. PMLR (2021)
Jorge, J., Vieco, J., Paredes, R., Sánchez, J., Benedí, J.: Empirical evaluation of variational autoencoders for data augmentation. In: Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2018), Volume 5: VISAPP, Funchal, Madeira, Portugal, 27–29 January 2018, pp. 96–104. SciTePress (2018)
Antoniou, A., Storkey, A., Edwards, H.: Augmenting image classifiers using data augmentation generative adversarial networks. In: Kůrková, V., Manolopoulos, Y., Hammer, B., Iliadis, L., Maglogiannis, I. (eds.) ICANN 2018. LNCS, vol. 11141, pp. 594–603. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01424-7_58
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. In: 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, 7–9 May 2015. Conference Track Proceedings (2015)
Kim, I., Kim, Y., Kim, S.: Learning loss for test-time augmentation. In: Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, Virtual, 6–12 December 2020 (2020)
Shanmugam, D., Blalock, D.W., Balakrishnan, G., Guttag, J.V.: Better aggregation in test-time augmentation. In: 2021 IEEE/CVF International Conference on Computer Vision, ICCV 2021, Montreal, QC, Canada, 10–17 October 2021, pp. 1194–1203. IEEE (2021)
Park, E., et al.: Big/little deep neural network for ultra low power inference. In: 2015 International Conference on Hardware/Software Codesign and System Synthesis, CODES+ISSS 2015, Amsterdam, Netherlands, 4–9 October 2015, pp. 124–132. IEEE (2015)
Mocerino, L., Calimera, A.: Fast and accurate inference on microcontrollers with boosted cooperative convolutional neural networks (BC-Net). IEEE Trans. Circuits Syst. I Regul. Pap. 68(1), 77–88 (2020)
Joshi, A.J., Porikli, F., Papanikolopoulos, N.: Multi-class active learning for image classification. In: 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009), Miami, Florida, USA, 20–25 June 2009, pp. 2372–2379. IEEE Computer Society (2009)
Wang, K., Yan, X., Zhang, D., Zhang, L., Lin, L.: Towards human-machine cooperation: Self-supervised sample mining for object detection. In: 2018 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2018, Salt Lake City, UT, USA, 18–22 June 2018, pp. 1605–1613. Computer Vision Foundation/IEEE Computer Society (2018)
Rizzo, R.G., Peluso, V., Calimera, A.: TVFS: topology voltage frequency scaling for reliable embedded convnets. IEEE Trans. Circuits Syst. II Express Briefs 68(2), 672–676 (2020)
Shannon, C.E.: A mathematical theory of communication. ACM SIGMOBILE Mob. Comput. Commun. Rev. 5(1), 3–55 (2001)
Melacci, S., Gori, M.: Unsupervised learning by minimal entropy encoding. IEEE Trans. Neural Netw. Learn. Syst. 23(12), 1849–1861 (2012)
Hassibi, B., Shadbakht, S.: Normalized entropy vectors, network information theory and convex optimization. In: Proceedings of the IEEE Information Theory Workshop on Information Theory for Wireless Networks, Solstrand, Norway, 1–6 July 2007, pp. 1–5. IEEE (2007)
Linaro toolchain. https://www.linaro.org/downloads/
Tensorflow Hub. https://tfhub.dev
Tensorflow lite hosted models. https://www.tensorflow.org/lite/guide/hosted_models
Howard, A.G., et al.: MobileNets: efficient convolutional neural networks for mobile vision applications. CoRR, vol. abs/1704.04861 (2017)
Sandler, M., Howard, A.G., Zhu, M., Zhmoginov, A., Chen, L.: MobileNetV2: inverted residuals and linear bottlenecks. In: 2018 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2018, Salt Lake City, UT, USA, 18–22 June 2018, pp. 4510–4520. Computer Vision Foundation/IEEE Computer Society (2018)
Howard, A., et al.: Searching for MobileNetV3. In: 2019 IEEE/CVF International Conference on Computer Vision, ICCV 2019, Seoul, Korea (South), 27 October–2 November 2019, pp. 1314–1324. IEEE (2019)
Deng, J., et al.: ImageNet: a large-scale hierarchical image database. In: 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009), Miami, Florida, USA, 20–25 June 2009, pp. 248–255. IEEE Computer Society (2009)
Peluso, V., Rizzo, R.G., Calimera, A., Macii, E., Alioto, M.: Beyond ideal DVFS through ultra-fine grain Vdd-hopping. In: Hollstein, T., Raik, J., Kostin, S., Tšertov, A., O’Connor, I., Reis, R. (eds.) VLSI-SoC 2016. IAICT, vol. 508, pp. 152–172. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-67104-8_8
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 IFIP International Federation for Information Processing
About this paper
Cite this paper
Mocerino, L., Rizzo, R.G., Peluso, V., Calimera, A., Macii, E. (2022). On the Efficiency of AdapTTA: An Adaptive Test-Time Augmentation Strategy for Reliable Embedded ConvNets. In: Grimblatt, V., Chang, C.H., Reis, R., Chattopadhyay, A., Calimera, A. (eds) VLSI-SoC: Technology Advancement on SoC Design. VLSI-SoC 2021. IFIP Advances in Information and Communication Technology, vol 661. Springer, Cham. https://doi.org/10.1007/978-3-031-16818-5_1
Download citation
DOI: https://doi.org/10.1007/978-3-031-16818-5_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-16817-8
Online ISBN: 978-3-031-16818-5
eBook Packages: Computer ScienceComputer Science (R0)