Document Open Access Logo

Faster Classification of Time-Series Input Streams

Authors Kunal Agrawal , Sanjoy Baruah , Zhishan Guo , Jing Li , Federico Reghenzani , Kecheng Yang , Jinhao Zhao

PDF HTML 5

Files

Thumbnail PDF
PDF
LIPIcs.ECRTS.2025.13.pdf
  • Filesize: 1.36 MB
  • 22 pages
HTML5 Logo HTML (experimental)
LIPIcs.ECRTS.2025.13.html

Document Identifiers

Subject Classification

ACM Subject Classification
  • Computer systems organization → Real-time systems
Keywords
  • Classification
  • Deep Learning
  • Sensor data streams
  • IDK classifiers

Metrics

  • Access Statistics
  • Total Accesses (updated on a weekly basis)
    0
    PDF Downloads
    0
    Metadata Views

Abstract

Deep learning–based classifiers are widely used for perception in autonomous Cyber-Physical Systems (CPS’s). However, such classifiers rarely offer guarantees of perfect accuracy while being optimized for efficiency. To support safety-critical perception, ensembles of multiple different classifiers working in concert are typically used. Since CPS’s interact with the physical world continuously, it is not unreasonable to expect dependencies among successive inputs in a stream of sensor data. Prior work introduced a classification technique that leverages these inter-input dependencies to reduce the average time to successful classification using classifier ensembles. In this paper, we propose generalizations to this classification technique, both in the improved generation of classifier cascades and the modeling of temporal dependencies. We demonstrate, through theoretical analysis and numerical evaluation, that our approach achieves further reductions in average classification latency compared to the prior methods.

Cite As Get BibTex

Kunal Agrawal, Sanjoy Baruah, Zhishan Guo, Jing Li, Federico Reghenzani, Kecheng Yang, and Jinhao Zhao. Faster Classification of Time-Series Input Streams. In 37th Euromicro Conference on Real-Time Systems (ECRTS 2025). Leibniz International Proceedings in Informatics (LIPIcs), Volume 335, pp. 13:1-13:22, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025) https://doi.org/10.4230/LIPIcs.ECRTS.2025.13

Author Details

Kunal Agrawal
  • Washington University in Saint Louis, MO, USA
Sanjoy Baruah
  • Washington University in Saint Louis, MO, USA
Zhishan Guo
  • North Carolina State University, Raleigh, NC, USA
Jing Li
  • New Jersey Institute of Technology, Newark, NJ, USA
Federico Reghenzani
  • Politecnico di Milano, Italy
Kecheng Yang
  • Texas State University, San Marcos, TX, USA
Jinhao Zhao
  • Washington University in Saint Louis, MO, USA

Funding

  • Agrawal, Kunal: NSF Grant No. CCF-2106699, CCF-2107280, and PPoSS-2216971.
  • Baruah, Sanjoy: NSF Grant No. CNS-2141256 and CPS-2229290.
  • Guo, Zhishan: NSF Grant No. CMMI-2246672.
  • Li, Jing: NSF Grant No. CNS-2340171, DOE ASCR Award Number DE-SC0024424.
  • Reghenzani, Federico: NGI Enrichers Transatlantic Fellowship Programme, National Resilience and Recovery Plan (PNRR) through the National Center for HPC, Big Data and Quantum Computing.
  • Yang, Kecheng: NSF Grant No. CNS-2104181.

Supplementary Materials

References

  1. Tarek Abdelzaher, Kunal Agrawal, Sanjoy Baruah, Alan Burns, Robert I. Davis, Zhishan Guo, and Yigong Hu. Scheduling IDK classifiers with arbitrary dependences to minimize the expected time to successful classification. Real-Time Systems, March 2023. URL: https://doi.org/10.1007/s11241-023-09395-0.
  2. Kunal Agrawal, Sanjoy Baruah, Alan Burns, and Jinhao Zhao. IDK cascades for time-series input streams. In Proceedings of the IEEE Real-Time Systems Symposium (RTSS), York, UK, 2024. IEEE Computer Society Press. Google Scholar
  3. Sanjoy Baruah, Iain Bate, Alan Burns, and Robert Davis. Optimal synthesis of fault-tolerant IDK cascades for real-time classification. In Proceedings of the 30th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS 2024). IEEE, 2024. URL: https://doi.org/10.1109/RTAS61025.2024.00011.
  4. Sanjoy Baruah, Alan Burns, Robert Davis, and Yue Wu. Optimally ordering IDK classifiers subject to deadlines. Real Time Syst., 2022. URL: https://doi.org/10.1007/s11241-022-09383-w.
  5. Sanjoy Baruah, Alan Burns, and Robert I. Davis. Optimal synthesis of robust IDK classifier cascades. In Claire Pagetti and Alessandro Biondi, editors, 2023 International Conference on Embedded Software, EMSOFT 2023, Hamburg, Germany, September 2023. ACM, 2023. URL: https://doi.org/10.1145/3609129.
  6. Sanjoy Baruah, Alan Burns, and Yue Wu. Optimal synthesis of IDK-cascades. In Proceedings of the Twenty-Ninth International Conference on Real-Time and Network Systems, RTNS '21, New York, NY, USA, 2021. ACM. URL: https://doi.org/10.1145/3453417.3453425.
  7. Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest, and Clifford Stein. Introduction to Algorithms, 3rd Edition. MIT Press, 2009. Google Scholar
  8. Yizeng Han, Gao Huang, Shiji Song, Le Yang, Honghui Wang, and Yulin Wang. Dynamic neural networks: A survey. IEEE transactions on pattern analysis and machine intelligence, 44(11):7436-7456, 2021. URL: https://doi.org/10.1109/TPAMI.2021.3117837.
  9. Yigong Hu, Shengzhong Liu, Tarek Abdelzaher, Maggie Wigness, and Philip David. On exploring image resizing for optimizing criticality-based machine perception. In 2021 IEEE 27th International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA), pages 169-178, 2021. URL: https://doi.org/10.1109/RTCSA52859.2021.00027.
  10. Metod Jazbec, James Allingham, Dan Zhang, and Eric Nalisnick. Towards anytime classification in early-exit architectures by enforcing conditional monotonicity. Advances in Neural Information Processing Systems, 36:56138-56168, 2023. Google Scholar
  11. Wittawat Jitkrittum, Neha Gupta, Aditya K Menon, Harikrishna Narasimhan, Ankit Rawat, and Sanjiv Kumar. When does confidence-based cascade deferral suffice? Advances in Neural Information Processing Systems, 36:9891-9906, 2023. Google Scholar
  12. Yigitcan Kaya, Sanghyun Hong, and Tudor Dumitras. Shallow-deep networks: Understanding and mitigating network overthinking. In International conference on machine learning, pages 3301-3310. PMLR, 2019. URL: http://proceedings.mlr.press/v97/kaya19a.html.
  13. In Jae Myung. Tutorial on maximum likelihood estimation. Journal of Mathematical Psychology, 47(1):90-100, 2003. URL: https://doi.org/10.1016/S0022-2496(02)00028-7.
  14. Ishrak Jahan Ratul, Zhishan Guo, and Kecheng Yang. Cascading idk classifiers to accelerate object recognition while preserving accuracy. In 49th IEEE International Conference on Computers, Software, and Applications (COMPSAC). IEEE, 2025. Google Scholar
  15. Xin Wang, Yujia Luo, Daniel Crankshaw, Alexey Tumanov, and Joseph E. Gonzalez. IDK cascades: Fast deep learning by learning not to overthink. CoRR, abs/1706.00885, 2017. URL: https://arxiv.org/abs/1706.00885.
Any Issues?
X

Feedback on the Current Page

CAPTCHA

Thanks for your feedback!

Feedback submitted to Dagstuhl Publishing

Could not send message

Please try again later or send an E-mail
© 2023-2025 Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact