Predictive Inference with Feature Conformal Prediction

Teng, Jiaye; Wen, Chuang; Zhang, Dinghuai; Bengio, Yoshua; Gao, Yang; Yang, Yuan

doi:10.48550/arxiv.2210.00173

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, to tackle the uncertainty issues, traditional methods (e.g., Bayesian approximation) for establishing confidence in prediction models usually make strict assumptions and have high computational cost properties. In this work, we build upon conformal prediction (Gibbs and Candes 2021;Chernozhukov, Wüthrich, and Yinchu 2018;Xu and Xie 2023;Tibshirani et al 2019;Teng et al 2022;Martinez et al 2023;Ndiaye 2022;Lu et al 2022;Jaramillo and Smirnov 2021;Liu et al 2022), which do not require strict assumptions on the underlying data distribution and has been applied before to classification and regression problems to attain marginal coverage. However, our work is different from existing conformal works in that we first model explanation prediction uncertainties in both scenarios -with and without true explanations.…”

Section: Related Workmentioning

confidence: 99%

Towards Modeling Uncertainties of Self-Explaining Neural Networks via Conformal Prediction

Qian,

Zhao,

et al. 2024

AAAI

View full text Add to dashboard Cite

Despite the recent progress in deep neural networks (DNNs), it remains challenging to explain the predictions made by DNNs. Existing explanation methods for DNNs mainly focus on post-hoc explanations where another explanatory model is employed to provide explanations. The fact that post-hoc methods can fail to reveal the actual original reasoning process of DNNs raises the need to build DNNs with built-in interpretability. Motivated by this, many self-explaining neural networks have been proposed to generate not only accurate predictions but also clear and intuitive insights into why a particular decision was made. However, existing self-explaining networks are limited in providing distribution-free uncertainty quantification for the two simultaneously generated prediction outcomes (i.e., a sample's final prediction and its corresponding explanations for interpreting that prediction). Importantly, they also fail to establish a connection between the confidence values assigned to the generated explanations in the interpretation layer and those allocated to the final predictions in the ultimate prediction layer. To tackle the aforementioned challenges, in this paper, we design a novel uncertainty modeling framework for self-explaining networks, which not only demonstrates strong distribution-free uncertainty modeling performance for the generated explanations in the interpretation layer but also excels in producing efficient and effective prediction sets for the final predictions based on the informative high-level basis explanations. We perform the theoretical analysis for the proposed framework. Extensive experimental evaluation demonstrates the effectiveness of the proposed uncertainty framework.

show abstract