arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2209.09056cs.LGcs.AI
收藏

概念嵌入模型:超越准确性与可解释性之间的权衡

Concept Embedding Models: Beyond the Accuracy-Explainability Trade-Off

University of Cambridge(剑桥大学);Inria(法国国家信息与自动化研究所);CNRS(法国国家科学研究中心);KU Leuven(鲁汶大学);University of Siena(锡耶纳大学);Babylon Health(巴比伦健康公司);Alan Turing Institute(艾伦·图灵研究所)
Mateo Espinosa Zarlenga, Pietro Barbiero, Gabriele Ciravegna, Giuseppe Marra, Francesco Giannini, Michelangelo Diligenti, Zohreh Shams, Frederic Precioso, Stefa… 展开作者

Mateo Espinosa Zarlenga, Pietro Barbiero, Gabriele Ciravegna, Giuseppe Marra, Francesco Giannini, Michelangelo Diligenti, Zohreh Shams, Frederic Precioso, Stefano Melacci, Adrian Weller, Pietro Lio, Mateja Jamnik

AI总结:

本文提出Concept Embedding Models,通过学习可解释的高维概念表示,在任务准确性、概念解释和测试时干预之间突破传统权衡。

摘要:

部署由AI驱动的系统需要可信赖的模型来支持有效的人机交互,而不能只关注原始预测准确性。概念瓶颈模型通过将分类任务建立在类人概念的中间层之上,提升了可信赖性。这使得人类干预成为可能,即可以纠正预测错误的概念,从而提高模型性能。然而,现有的概念瓶颈模型无法在高任务准确性、稳健的基于概念的解释以及有效的概念干预之间找到最优折中,尤其是在完整且准确的概念监督稀缺的真实条件下。为解决这一问题,我们提出了Concept Embedding Models(概念嵌入模型),这是一类新颖的概念瓶颈模型,通过学习可解释的高维概念表示,超越了当前准确性与可解释性之间的权衡。实验表明,Concept Embedding Models:(1)相对于不使用概念的标准神经模型,取得了更好或具有竞争力的任务准确性;(2)提供了可捕获有意义语义的概念表示,这些语义包括且超出其真实标签;(3)支持测试时概念干预,并且其对测试准确性的影响超过标准概念瓶颈模型;(4)能够扩展到完整概念监督稀缺的真实条件。

英文摘要:

Deploying AI-powered systems requires trustworthy models supporting effective human interactions, going beyond raw prediction accuracy. Concept bottleneck models promote trustworthiness by conditioning classification tasks on an intermediate level of human-like concepts. This enables human interventions which can correct mispredicted concepts to improve the model's performance. However, existing concept bottleneck models are unable to find optimal compromises between high task accuracy, robust concept-based explanations, and effective interventions on concepts -- particularly in real-world conditions where complete and accurate concept supervisions are scarce. To address this, we propose Concept Embedding Models, a novel family of concept bottleneck models which goes beyond the current accuracy-vs-interpretability trade-off by learning interpretable high-dimensional concept representations. Our experiments demonstrate that Concept Embedding Models (1) attain better or competitive task accuracy w.r.t. standard neural models without concepts, (2) provide concept representations capturing meaningful semantics including and beyond their ground truth labels, (3) support test-time concept interventions whose effect in test accuracy surpasses that in standard concept bottleneck models, and (4) scale to real-world conditions where complete concept supervisions are scarce.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑