Knowing when we do not know: Bayesian continual learning for sensing-based analysis tasks

Despite much research targeted at enabling conventional machine learning models to continually learn tasks and data distributions sequentially without forgetting the knowledge acquired, little effort has been devoted to account for more realistic situations where learning some tasks accurately might be more critical than forgetting previous ones. In this paper we propose a Bayesian inference based framework to continually learn a set of real-world, sensing-based analysis tasks that can be tuned to prioritize the remembering of previously learned tasks or the learning of new ones. Our experiments prove the robustness and reliability of the learned models to adapt to the changing sensing environment, and show the suitability of using uncertainty of the predictions to assess their reliability.

[1]  Max Welling,et al.  Auto-Encoding Variational Bayes , 2013, ICLR.

[2]  Yarin Gal,et al.  Uncertainty in Deep Learning , 2016 .

[3]  Tommi S. Jaakkola,et al.  Learning Sleep Stages from Radio Signals: A Conditional Adversarial Architecture , 2017, ICML.

[4]  Julien Cornebise,et al.  Weight Uncertainty in Neural Network , 2015, ICML.

[5]  Gary M. Weiss,et al.  Activity recognition using cell phone accelerometers , 2011, SKDD.

[6]  Cecilia Mascolo,et al.  DSP.Ear: leveraging co-processor support for continuous audio sensing on smartphones , 2014, SenSys.

[7]  Georg Heigold,et al.  Small-footprint keyword spotting using deep neural networks , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[8]  Razvan Pascanu,et al.  Overcoming catastrophic forgetting in neural networks , 2016, Proceedings of the National Academy of Sciences.

[9]  Philip H. S. Torr,et al.  Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence , 2018, ECCV.

[10]  Ariel D. Procaccia,et al.  Variational Dropout and the Local Reparameterization Trick , 2015, NIPS.

[11]  Plácido Luna.  Free Sound Effects , 2016 .

[12]  George Tzanetakis,et al.  Musical genre classification of audio signals , 2002, IEEE Trans. Speech Audio Process..

[13]  Marc'Aurelio Ranzato,et al.  Gradient Episodic Memory for Continual Learning , 2017, NIPS.

[14]  Tarek F. Abdelzaher,et al.  FastDeepIoT: Towards Understanding and Optimizing Neural Network Execution Time on Mobile and Embedded Devices , 2018, SenSys.

[15]  Jiwon Kim,et al.  Continual Learning with Deep Generative Replay , 2017, NIPS.

[16]  Surya Ganguli,et al.  Continual Learning Through Synaptic Intelligence , 2017, ICML.

[17]  Geoffrey E. Hinton,et al.  Bayesian Learning for Neural Networks , 1995 .

[18]  Andre Wibisono,et al.  Streaming Variational Bayes , 2013, NIPS.

[19]  Alex Kendall,et al.  What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision? , 2017, NIPS.

[20]  Cecilia Mascolo,et al.  EmotionSense: a mobile phones based adaptive platform for experimental social psychology research , 2010, UbiComp.

[21]  Richard E. Turner,et al.  Variational Continual Learning , 2017, ICLR.

[22]  Xiaohui Peng,et al.  Deep Learning for Sensor-based Activity Recognition: A Survey , 2017, Pattern Recognit. Lett..

[23]  Daniel Gatica-Perez,et al.  StressSense: detecting stress in unconstrained acoustic environments using smartphones , 2012, UbiComp.

[24]  Christoph H. Lampert,et al.  iCaRL: Incremental Classifier and Representation Learning , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[25]  Samy Bengio,et al.  Large-scale content-based audio retrieval from text queries , 2008, MIR '08.

[26]  Alex Graves,et al.  Practical Variational Inference for Neural Networks , 2011, NIPS.

[27]  D LaneNicholas,et al.  Low-resource Multi-task Audio Sensing for Mobile and Embedded Devices via Shared Deep Neural Network Representations , 2017 .

[28]  Erik McDermott,et al.  Deep neural networks for small footprint text-dependent speaker verification , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).