Interpretable by Design: Learning Predictors by Composing Interpretable Queries
Interpretable by Design: Learning Predictors by Composing Interpretable Queries
复制标题
DOI：
10.1109/tpami.2022.3225162
复制
发表时间：
2022-07
影响因子：
23.6
通讯作者：
Aditya Chattopadhyay;Stewart Slocum;B. Haeffele;René Vidal;D. Geman
中科院分区：
文献类型：
--
作者：
Aditya Chattopadhyay;Stewart Slocum;B. Haeffele;René Vidal;D. Geman
人们越来越担心使用高性能机器学习算法进行典型的不透明决策。以特定于领域的术语解释推理过程对于在医疗保健等风险敏感领域采用可能是至关重要的。我们认为，机器学习算法应该是可设计解释的，表达这些解释的语言应该是领域和任务相关的。因此，我们的模型的预测基于一系列用户定义的和特定于任务的数据二进制函数，每个函数都对最终用户有明确的解释。然后，我们最大限度地减少对任何给定输入进行准确预测所需的预期查询数。由于解决方案一般很难解决，因此，在前面的工作基础上，我们基于信息增益顺序地选择查询。然而，与以前的工作不同，我们不需要假设查询是条件独立的。相反，我们利用随机生成模型(VAE)和MCMC算法(未调整的朗之万算法)来根据先前的查询-答案来选择关于输入的最具信息量的查询。这使得能够在线确定解决预测歧义所需的任何深度的查询链。最后，在视觉任务和自然语言处理任务上的实验证明了该方法的有效性及其相对于后解释的优越性。
There is a growing concern about typically opaque decision-making with high-performance machine learning algorithms. Providing an explanation of the reasoning process in domain-specific terms can be crucial for adoption in risk-sensitive domains such as healthcare. We argue that machine learning algorithms should be interpretable by design and that the language in which these interpretations are expressed should be domain- and task-dependent. Consequently, we base our model's prediction on a family of user-defined and task-specific binary functions of the data, each having a clear interpretation to the end-user. We then minimize the expected number of queries needed for accurate prediction on any given input. As the solution is generally intractable, following prior work, we choose the queries sequentially based on information gain. However, in contrast to previous work, we need not assume the queries are conditionally independent. Instead, we leverage a stochastic generative model (VAE) and an MCMC algorithm (Unadjusted Langevin) to select the most informative query about the input based on previous query-answers. This enables the online determination of a query chain of whatever depth is required to resolve prediction ambiguities. Finally, experiments on vision and NLP tasks demonstrate the efficacy of our approach and its superiority over post-hoc explanations.