paper-with-me

홈 › Papers

Unaligned but Safe -- Formally Compensating Performance Limitations for Imprecise 2D Object Detection

2022-02-10 · Tobias Schuster, Emmanouil Seferis, Simon Burton, Chih-Hong Cheng

In this paper, we consider the imperfection within machine learning-based 2D object detection and its impact on safety. We address a special sub-type of performance limitations: the prediction bounding box cannot be perfectly aligned with the ground truth, but the computed Intersection-over-Union metric is always larger than a given threshold. Under such type of performance limitation, we formally prove the minimum required bounding box enlargement factor to cover the ground truth. We then demonstrate that the factor can be mathematically adjusted to a smaller value, provided that the motion planner takes a fixed-length buffer in making its decisions. Finally, observing the difference between an empirically measured enlargement factor and our formally derived worst-case enlargement factor offers an interesting connection between the quantitative evidence (demonstrated by statistics) and the qualitative evidence (demonstrated by worst-case analysis).

📄 PDF Abstract BibTeX arXiv:2202.05123

Code (0)

등록된 구현이 없습니다.

Tasks

2D Object Detectionobject-detectionObject Detection

Similar Papers 제목 키워드 기반

Stealthy and Persistent Unalignment on Large Language Models via Backdoor Injections

2023-11-15 · Yuanpu Cao, Bochuan Cao, Jinghui Chen

Recent developments in Large Language Models (LLMs) have manifested significant advancements. To facilitate safeguards against malicious exploitation, a body of research has concentrated on aligning LLMs with human prefe…

Red Teaming

CRSOT: Cross-Resolution Object Tracking using Unaligned Frame and Event Cameras

2024-01-05 · Yabin Zhu, Xiao Wang, Chenglong Li, Bo Jiang 외

Existing datasets for RGB-DVS tracking are collected with DVS346 camera and their resolution ($346 \times 260$) is low for practical applications. Actually, only visible cameras are deployed in many practical systems, an…

Object Tracking

Unpredictability of AI

2019-05-29 · Roman V. Yampolskiy

The young field of AI Safety is still in the process of identifying its challenges and limitations. In this paper, we formally describe one such impossibility result, namely Unpredictability of AI. We prove that it is im…

Decoupled Alignment for Robust Plug-and-Play Adaptation

2024-06-03 · Haozheng Luo, Jiahao Yu, Wenxin Zhang, Jialong Li 외

We introduce a low-resource safety enhancement method for aligning large language models (LLMs) without the need for supervised fine-tuning (SFT) or reinforcement learning from human feedback (RLHF). Our main idea is to …

Knowledge Distillation

Immersion and Invariance-based Disturbance Observer and Its Application to Safe Control

2023-09-13 · Yujie Wang, Xiangru Xu

When the disturbance input matrix is nonlinear, existing disturbance observer design methods rely on the solvability of a partial differential equation or the existence of an output function with a uniformly well-defined…