Papers
Topics
Authors
Recent
Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
Gemini 2.5 Flash
Gemini 2.5 Flash 172 tok/s
Gemini 2.5 Pro 48 tok/s Pro
GPT-5 Medium 30 tok/s Pro
GPT-5 High 29 tok/s Pro
GPT-4o 103 tok/s Pro
Kimi K2 199 tok/s Pro
GPT OSS 120B 464 tok/s Pro
Claude Sonnet 4.5 36 tok/s Pro
2000 character limit reached

System-Level Analysis of Module Uncertainty Quantification in the Autonomy Pipeline (2410.12019v1)

Published 15 Oct 2024 in eess.SY and cs.SY

Abstract: We present a novel perspective on the design, use, and role of uncertainty measures for learned modules in an autonomous system. While in the current literature uncertainty measures are produced for standalone modules without considering the broader system context, in our work we explicitly consider the role of decision-making under uncertainty in illuminating how "good'" an uncertainty measure is. Our insights are centered around quantifying the ways in which being uncertainty-aware makes a system more robust. Firstly, we use level set generation tools to produce a measure for system robustness and use this measure to compare system designs, thus placing uncertainty quantification in the context of system performance and evaluation metrics. Secondly, we use the concept of specification generation from systems theory to produce a formulation under which a designer can simultaneously constrain the properties of an uncertainty measure and analyze the efficacy of the decision-making-under-uncertainty algorithm used by the system. We apply our analyses to two real-world and complex autonomous systems, one for autonomous driving and another for aircraft runway incursion detection, helping to form a toolbox for an uncertainty-aware system designer to produce more effective and robust systems.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (41)
  1. P. Akella and A. D. Ames. Disturbance bounds for signal temporal logic task satisfaction: A dynamics perspective. IEEE Control Systems Letters, pages 2018–2023, 2021.
  2. Differentiable MPC for end-to-end planning and control. In Advances in Neural Information Processing Systems, pages 8299–8310, 2018.
  3. System level synthesis. Annual Reviews in Control, 47, 05 2019.
  4. Uncertainty sets for image classifiers using conformal prediction. arXiv, 2020.
  5. A. N. Angelopoulos and S. Bates. A gentle introduction to conformal prediction and distribution-free uncertainty quantification. arXiv preprint 2107.07511, 2021.
  6. nuScenes: A multimodal dataset for autonomous driving. In Conference on Computer Vision and Pattern Recognition, 2020.
  7. Evora: Deep evidential traversability learning for risk-aware off-road autonomy. ArXiv, abs/2311.06234, 2023.
  8. M. C. Campi and S. Garatti. Scenario optimization with relaxation: a new tool for design and application to machine learning problems. In 2020 59th IEEE Conference on Decision and Control (CDC), pages 2463–2468, 2020.
  9. K. Chakraborty and S. Bansal. Discovering closed-loop failures of vision-based controllers via reachability analysis. IEEE Robotics and Automation Letters, PP:1–8, 05 2023.
  10. Disentangling epistemic and aleatoric uncertainty in reinforcement learning. In CoRR, 2022.
  11. Risk-driven design of perception systems. ArXiv, abs/2205.10677, 2022.
  12. C. Dawson and C. Fan. Certifiable robot design optimization using differentiable programming. In Robotics: Science and Systems, 2022.
  13. C. Dawson and C. Fan. A bayesian approach to breaking things: Efficiently predicting and repairing failure modes via sampling. In Conference on Robot Learning, 2023.
  14. S. Dean. Reliable Machine Learning in Feedback Systems. PhD thesis, EECS Department, University of California, Berkeley, Aug 2021.
  15. P. Frazier. A tutorial on bayesian optimization. ArXiv, abs/1807.02811, 2018.
  16. Y. Gal and Z. Ghahramani. Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In M. F. Balcan and K. Q. Weinberger, editors, Proceedings of The 33rd International Conference on Machine Learning, volume 48 of Proceedings of Machine Learning Research, pages 1050–1059, New York, New York, USA, 20–22 Jun 2016. PMLR.
  17. S. Garatti and M. Campi. On the consistency of the risk evaluation in the scenario approach. pages 1468–1473, 12 2021.
  18. Active learning for level set estimation. IJCAI International Joint Conference on Artificial Intelligence, pages 1344–1350, 08 2013.
  19. Active learning for distributionally robust level-set estimation. 2021.
  20. Pacti: Scaling assume-guarantee reasoning for system analysis and design. ArXiv, abs/2303.17751, 2023.
  21. B. Ivanovic and M. Pavone. Injecting planning-awareness into prediction and detection evaluation. In IEEE Intelligent Vehicles Symposium, 2022.
  22. Bayesian experimental design for finding reliable level set under input uncertainty. IEEE Access, 8:203982–203993, 01 2020.
  23. Diffstack: A differentiable and modular control stack for autonomous vehicles. In 6th Annual Conference on Robot Learning, 2022.
  24. Efficient determination of safety requirements for perception systems. In Digital Avionics Systems Conference, 2023.
  25. Multivariate confidence calibration for object detection. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 1322–1330, 2020.
  26. Simple and scalable predictive uncertainty estimation using deep ensembles. In Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS’17, page 6405–6416, Red Hook, NY, USA, 2017. Curran Associates Inc.
  27. Conformal decision theory: Safe autonomous decisions without distributions. arXiv, 2024.
  28. A general framework for uncertainty estimation in deep learning. IEEE Robotics and Automation Letters, PP:1–1, 02 2020.
  29. A quantitative framework for layered multirate control: Toward a theory of control architecture. IEEE Control Systems, 44:52–94, Jun 2024.
  30. Multi-agent reachability calibration with conformal prediction. pages 6596–6603, 12 2023.
  31. The complexity of Markov decision processes. Mathematics of Operations Research, 12(3):441–450, 1987.
  32. Sampling-free epistemic uncertainty estimation using approximated variance propagation. 2019 IEEE/CVF International Conference on Computer Vision (ICCV), pages 2931–2940, 2019.
  33. Trajectron++: Dynamically-feasible trajectory forecasting with heterogeneous data. In European Conference on Computer Vision, 2020.
  34. Sketching curvature for efficient out-of-distribution detection for deep neural networks. ArXiv, abs/2102.12567, 2021.
  35. Fcos: Fully convolutional one-stage object detection. pages 9626–9635, 10 2019.
  36. A. Wald. Sequential Analysis. John Wiley and Sons, New York, 1947.
  37. Quality of uncertainty quantification for bayesian neural network inference. arXiv preprint 1906.09686, 2019.
  38. Robust super-level set estimation using gaussian processes. In ECML/PKDD, 2018.
  39. G. Zardini. Co-Design of Complex Systems: From Autonomy to Future Mobility Systems. PhD thesis, ETH Zurich, 2023.
  40. Why change your controller when you can change your planner: Drag-aware trajectory generation for quadrotor systems. ArXiv, abs/2401.04960, 2024.
  41. Theoretically principled trade-off between robustness and accuracy. In International conference on machine learning, pages 7472–7482. PMLR, 2019.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Lightbulb Streamline Icon: https://streamlinehq.com

Continue Learning

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now
List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up