In the field of finance, the underwriting process is an essential step in evaluating every loan application. During this stage, the borrowers' creditworthiness and ability to repay the loan are assessed to ultimately decide whether to approve the loan application. One of the core components of underwriting is credit scoring, in which the probability of default is estimated. As such, there has been significant progress in enhancing the predictive accuracy of credit scoring models through the use of machine learning, but there still exists a need to ultimately construct an approval rule that takes into consideration additional criteria beyond the score itself. This construction process is traditionally done manually to ensure that the approval rule remains interpretable to humans. In this paper, we outline an automated system for optimizing a rule-based system for approving loan applications, which has been deployed at Hyundai Capital Services (HCS). The main challenge lay in creating a high-quality rule base that is simultaneously simple enough to be interpretable by risk analysts as well as customers, since the approval decision should be accountable. We addressed this challenge through principled submodular optimization. The deployment of our system has led to a 14% annual growth in the volume of loan services at HCS, while maintaining the target bad rate, and has resulted in the approval of customers who might have otherwise been rejected.
论文检索
输入标题、作者或关键词，从 7,537 篇学术成果中精准定位
G–LIME: Statistical Learning for Local Interpretations of Deep Neural Networks Using Global Priors (Abstract Reprint)
PDF ↗To explain the prediction result of a Deep Neural Network (DNN) model based on a given sample, LIME [1] and its derivatives have been proposed to approximate the local behavior of the DNN model around the data point via linear surrogates. Though these algorithms interpret the DNN by finding the key features used for classification, the random interpolations used by LIME would perturb the explanation result and cause the instability and inconsistency between repetitions of LIME computations. To tackle this issue, we propose G-LIME that extends the vanilla LIME through high-dimensional Bayesian linear regression using the sparsity and informative global priors. Specifically, with a dataset representing the population of samples (e.g., the training set), G-LIME first pursues the global explanation of the DNN model using the whole dataset. Then, with a new data point, -LIME incorporates an modified estimator of ElasticNet-alike to refine the local explanation result through balancing the distance to the global explanation and the sparsity/feature selection in the explanation. Finally, G-LIME uses Least Angle Regression (LARS) and retrieves the solution path of a modified ElasticNet under varying -regularization, to screen and rank the importance of features [2] as the explanation result. Through extensive experiments on real world tasks, we show that the proposed method yields more stable, consistent, and accurate results compared to LIME.
Devising a strategy to make a system mimic behaviors from another system is a problem that naturally arises in many areas of Computer Science. In this work, we interpret this problem in the context of intelligent agents, from the perspective of LTLf, a formalism commonly used in AI for expressing finite-trace properties. Our model consists of two separated dynamic domains, D_A and D_B, and an LTLf specification that formalizes the notion of mimicking by mapping properties on behaviors (traces) of D_A into properties on behaviors of D_B. The goal is to synthesize a strategy that step-by-step maps every behavior of D_A into a behavior of D_B so that the specification is met. We consider several forms of mapping specifications, ranging from simple ones to full LTLf, and for each, we study synthesis algorithms and computational properties.
Relational structured data is a way of representing knowledge using nodes and edges, while also capturing the meaning of that knowledge in a structured form that can be used for machine learning. Compared with vision and natural language data, relational structured data represents and manipulates structured knowledge, which can be beneficial for tasks that involve reasoning or inference. On the other hand, vision and NLP deal more with unstructured data (like images and text), and they often require different types of models and algorithms to extract useful information or features from the data. Human-like Learning develops methods that can harness relational structures and learning-to-learn to rapidly acquire and generalize knowledge to new tasks and situations. With Human-like Learning, the learning algorithm is efficient and can adapt to new or unseen situations, which is crucial in real-world applications where environments may change unpredictably. Moreover, the models are easier for humans to understand and interpret, which is important for transparency and trust in AI systems. In this talk, we present our recent attempts towards human-like learning from relational structured data.
Deep neural networks (DNNs) have achieved unprecedented success across many scientific and engineering fields in the last decades. Despite its empirical success, unfortunately, recent studies have shown that there are various failure modes and blindspots in DNN models which may result in unexpected serious failures and potential harms, e.g. the existence of adversarial examples and small perturbations. This is not acceptable especially for safety critical and high stakes applications in the real-world, including healthcare, self-driving cars, aircraft control systems, hiring and malware detection protocols. Moreover, it has been challenging to understand why and when DNNs will fail due to their complicated structures and black-box behaviors. Lacking interpretability is one critical issue that may seriously hinder the deployment of DNNs in high-stake applications, which need interpretability to trust the prediction, to understand potential failures, and to be able to mitigate harms and eliminate biases in the model. To make DNNs trustworthy and reliable for deployment, it is necessary and urgent to develop methods and tools that can (i) quantify and improve their robustness against adversarial and natural perturbations, and (ii) understand their underlying behaviors and further correct errors to prevent injuries and damages. These are the important first steps to enable Trustworthy AI and Trustworthy Machine Learning. In this talk, I will survey a series of research efforts in my lab contributed to tackling the grand challenges in (i) and (ii). In the first part of my talk, I will overview our research effort in Robust Machine Learning since 2017, where we have proposed the first attack-agnostic robustness evaluation metric, the first efficient robustness certification algorithms for various types of perturbations, and efficient robust learning algorithms across supervised learning to deep reinforcement learning. In the second part of my talk, I will survey a series of exciting results in my lab on accelerating interpretable machine learning and explainable AI. Specifically, I will show how we could bring interpretability into deep learning by leveraging recent advances in multi-modal models. I'll present recent works in our group on automatically dissecting neural networks with open vocabulary concepts, designing interpretable neural networks without concept labels, and briefly overview our recent efforts on demystifying black-box DNN training process, automated neuron explanations for Large Language Models and the first robustness evaluation of a family of neuron-level interpretation techniques.
Inverse reinforcement learning (IRL) has seen significant advancements in recent years. This class of approaches aims to efficiently learn the underlying reward function that rationalizes the behavior exhibited by expert agents, often represented by humans. In contrast to mere behavioral cloning, the reconstruction of a reward function yields appealing implications, as it allows for more effective interpretability of the expert’s decisions and provides a transferable specification of the expert’s objectives for application in even different environments. Unlike the well-understood field of reinforcement learning (RL) from a theoretical perspective, IRL still grapples with limited understanding, significantly constraining its applicability. A fundamental challenge in IRL is the inherent ambiguity in selecting a reward function, given the existence of multiple candidate functions, all explaining the expert’s behavior. In this talk, I will survey three of my papers that have made notable contributions to the IRL field: “Provably Efficient Learning of Transferable Rewards”, “Towards Theoretical Understanding of Inverse Reinforcement Learning”, and “Inverse Reinforcement Learning with Sub-optimal Experts". The central innovation introduced by the first paper is a novel formulation of the IRL problem that overcomes the issue of ambiguity. IRL is reframed as the problem of learning the feasible reward set, which is the set of all rewards that can explain the expert’s behavior. This approach postpones the selection of the reward function, thereby circumventing the ambiguity issues. Furthermore, the feasible reward set exhibits convenient geometric properties that enable the development of efficient algorithms for its computation. Building on this novel formulation of IRL, the second paper addresses the problem of efficiently learning the feasible reward set when the environment and the expert’s policy are not known in advance. It introduces a novel way to assess the dissimilarity between feasible reward sets based on the Hausdorff distance and presents a new PAC (probabilistic approximately correct) framework. The most significant contribution of this paper is the introduction of the first sample complexity lower bound, which highlights the challenges inherent in the IRL problem. Deriving this lower bound necessitated the development of novel technical tools. The paper also demonstrates that when a generative model of the environment is available, a uniform sampling strategy achieves a sample complexity that matches the lower bound, up to logarithmic factors. Finally, in the third paper, the IRL problem in the presence of sub-optimal experts is investigated. Specifically, the paper assumes the availability of multiple sub-optimal experts, in addition to the expert agent, which provides additional demonstrations, associated with a known quantification of the maximum amount of sub-optimality. The paper shows that this richer information mitigates the ambiguity problem, significantly reducing the size of the feasible reward set while retaining its favorable geometric properties. Furthermore, the paper explores the associated statistical problem and derives novel lower bounds for sample complexity, along with almost matching algorithms. These selected papers represent notable advancements in IRL, contributing to the establishment of a solid theoretical foundation for IRL and extending the framework to accommodate scenarios with sub-optimal experts.
Neural models, including large language models (LLMs), achieve superior performance on logical reasoning tasks such as question answering. To elicit reasoning capabilities from LLMs, recent works propose using the chain-of-thought (CoT) mechanism to generate both the reasoning chain and the answer, which enhances the model’s capabilities in conducting reasoning. However, due to LLM’s uninterpretable nature and the extreme flexibility of free-form explanations, several challenges remain: such as struggling with inaccurate reasoning, hallucinations, and not aligning with human preferences. In this talk, we will focus on (1) our design of leveraging structured information (that is grounded to the context), for the explainable complex question answering and reasoning; (2) our multi-module interpretable framework for inductive reasoning, which conducts step-wise faithful reasoning with iterative feedback.
Recently, transformers have become incredibly popular in computer vision and vision-language tasks. This notable rise in their usage can be primarily attributed to the capabilities offered by attention mechanisms and the outstanding ability of transformers to adapt and apply themselves to a variety of tasks and domains. Their versatility and state-of-the-art performance have established them as indispensable tools for a wide array of applications. However, in the constantly changing landscape of machine learning, the assurance of the trustworthiness of transformers holds utmost importance. This paper conducts a thorough examination of vision-language transformers, employing three fundamental principles of responsible AI: Bias, Robustness, and Interpretability. The primary objective of this paper is to delve into the intricacies and complexities associated with the practical use of transformers, with the overarching goal of advancing our comprehension of how to enhance their reliability and accountability.
Evaluating Pre-trial Programs Using Interpretable Machine Learning Matching Algorithms for Causal Inference
PDF ↗After a person is arrested and charged with a crime, they may be released on bail and required to participate in a community supervision program while awaiting trial. These 'pre-trial programs' are common throughout the United States, but very little research has demonstrated their effectiveness. Researchers have emphasized the need for more rigorous program evaluation methods, which we introduce in this article. We describe a program evaluation pipeline that uses recent interpretable machine learning techniques for observational causal inference, and demonstrate these techniques in a study of a pre-trial program in Durham, North Carolina. Our findings show no evidence that the program either significantly increased or decreased the probability of new criminal charges. If these findings replicate, the criminal-legal system needs to either improve pre-trial programs or consider alternatives to them. The simplest option is to release low-risk individuals back into the community without subjecting them to any restrictions or conditions. Another option is to assign individuals to pre-trial programs that incentivize pro-social behavior. We believe that the techniques introduced here can provide researchers the rigorous tools they need to evaluate these programs.
Interpretable Long-Form Legal Question Answering with Retrieval-Augmented Large Language Models
PDF ↗Many individuals are likely to face a legal dispute at some point in their lives, but their lack of understanding of how to navigate these complex issues often renders them vulnerable. The advancement of natural language processing opens new avenues for bridging this legal literacy gap through the development of automated legal aid systems. However, existing legal question answering (LQA) approaches often suffer from a narrow scope, being either confined to specific legal domains or limited to brief, uninformative responses. In this work, we propose an end-to-end methodology designed to generate long-form answers to any statutory law questions, utilizing a "retrieve-then-read" pipeline. To support this approach, we introduce and release the Long-form Legal Question Answering (LLeQA) dataset, comprising 1,868 expert-annotated legal questions in the French language, complete with detailed answers rooted in pertinent legal provisions. Our experimental results demonstrate promising performance on automatic evaluation metrics, but a qualitative analysis uncovers areas for refinement. As one of the only comprehensive, expert-annotated long-form LQA dataset, LLeQA has the potential to not only accelerate research towards resolving a significant real-world issue, but also act as a rigorous benchmark for evaluating NLP models in specialized domains. We publicly release our code, data, and models.
Depression detection is a challenging and crucial task in psychological illness diagnosis. Utilizing online user posts to predict whether a user suffers from depression seems an effective and promising direction. However, existing methods suffer from either poor interpretability brought by the black-box models or underwhelming performance caused by the completely separate two-stage model structure. To alleviate these limitations, we propose a novel capsule network integrated with contrastive learning for depression detection (DeCapsNet). The highlights of DeCapsNet can be summarized as follows. First, it extracts symptom capsules from user posts by leveraging meticulously designed symptom descriptions, and then distills them into class-indicative depression capsules. The overall workflow is in an explicit hierarchical reasoning manner and can be well interpreted by the Patient Health Questionnaire-9 (PHQ9), which is one of the most widely adopted questionnaires for depression diagnosis. Second, it integrates with contrastive learning, which can facilitate the embeddings from the same class to be pulled closer, while simultaneously pushing the embeddings from different classes apart. In addition, by adopting the end-to-end training strategy, it does not necessitate additional data annotation, and mitigates the potential adverse effects from the upstream task to the downstream task. Extensive experiments on three widely-used datasets show that in both within-dataset and cross-dataset scenarios our proposed method outperforms other strong baselines significantly.
Accurate prediction of meteorological elements, such as temperature and relative humidity, is important to human livelihood, early warning of extreme weather, and urban governance. Recently, neural network-based methods have shown impressive performance in this field. However, most of them are overcomplicated and impenetrable. In this paper, we propose a straightforward and interpretable differential framework, where the key lies in explicitly estimating the evolutionary trends. Specifically, three types of trends are exploited. (1) The proximity trend simply uses the most recent changes. It works well for approximately linear evolution. (2) The sequential trend explores the global information, aiming to capture the nonlinear dynamics. Here, we develop an attention-based trend unit to help memorize long-term features. (3) The flow trend is motivated by the nature of evolution, i.e., the heat or substance flows from one region to another. Here, we design a flow-aware attention unit. It can reflect the interactions via performing spatial attention over flow maps. Finally, we develop a trend fusion module to adaptively fuse the above three trends. Extensive experiments on two datasets demonstrate the effectiveness of our method.
Predictive analytics has been widely used in various domains, including education, to inform decision-making and improve outcomes. However, many predictive models are proprietary and inaccessible for evaluation or modification by researchers and practitioners, limiting their accountability and ethical design. Moreover, predictive models are often opaque and incomprehensible to the officials who use them, reducing their trust and utility. Furthermore, predictive models may introduce or exacerbate bias and inequity, as they have done in many sectors of society. Therefore, there is a need for transparent, interpretable, and fair predictive models that can be easily adopted and adapted by different stakeholders. In this paper, we propose a fair predictive model based on multivariate adaptive regression splines (MARS) that incorporates fairness measures in the learning process. MARS is a non-parametric regression model that performs feature selection, handles non-linear relationships, generates interpretable decision rules, and derives optimal splitting criteria on the variables. Specifically, we integrate fairness into the knot optimization algorithm and provide theoretical and empirical evidence of how it results in a fair knot placement. We apply our fairMARS model to real-world data and demonstrate its effectiveness in terms of accuracy and equity. Our paper contributes to the advancement of responsible and ethical predictive analytics for social good.
This paper analyzes the privacy of traditional Statistical Disclosure Control (SDC) systems under a differential privacy interpretation. SDCs, such as cell suppression and swapping, promise to safeguard the confidentiality of data and are routinely adopted in data analyses with profound societal and economic impacts. Through a formal analysis and empirical evaluation of demographic data from real households in the U.S., the paper shows that widely adopted SDC systems not only induce vastly larger privacy losses than classical differential privacy mechanisms, but, they may also come at a cost of larger accuracy and fairness.
Line Probe Assay (LPA) is a widely used method for diagnosing drug-resistant tuberculosis (DRTB), but it is a time-consuming and labor-intensive process that requires expert interpretation. DRTB is a significant threat to global TB control efforts and its prompt diagnosis is critical for initiating appropriate treatment. In this paper, we present an automated LPA test interpretation solution that uses computer vision techniques to extract and analyze strips from LPA sheets and uses machine learning algorithms to produce drug sensitivity and resistivity outcomes with extremely high precision and recall. We also develop OCR models to eliminate manual data entry to further reduce the overall time. Our solution comprises a rejection module that flags ambiguous and novel samples that are then referred to experienced lab technicians. This results in increased trust in the solution. To evaluate our solution, we curate an extensive and diverse dataset of LPA strips annotated by multiple microbiologists across India. Our solution achieves more than 95% accuracy for all drugs on this dataset. The proposed solution has the potential to increase the efficiency, standardization of LPA test interpretation, and fast-tracking the dissemination of results to end-users via a designated Management Information System (MIS).
In recent years, there has been growing interest in developing robust machine learning (ML) models that can withstand adversarial attacks, including one of the most widely adopted, efficient, and interpretable ML algorithms—decision trees (DTs). This paper proposes a novel coevolutionary algorithm (CoEvoRDT) designed to create robust DTs capable of handling noisy high-dimensional data in adversarial contexts. Motivated by the limitations of traditional DT algorithms, we leverage adaptive coevolution to allow DTs to evolve and learn from interactions with perturbed input data. CoEvoRDT alternately evolves competing populations of DTs and perturbed features, enabling construction of DTs with desired properties. CoEvoRDT is easily adaptable to various target metrics, allowing the use of tailored robustness criteria such as minimax regret. Furthermore, CoEvoRDT has potential to improve the results of other state-of-the-art methods by incorporating their outcomes (DTs they produce) into the initial population and optimize them in the process of coevolution. Inspired by the game theory, CoEvoRDT utilizes mixed Nash equilibrium to enhance convergence. The method is tested on 20 popular datasets and shows superior performance compared to 4 state-of-the-art algorithms. It outperformed all competing methods on 13 datasets with adversarial accuracy metrics, and on all 20 considered datasets with minimax regret. Strong experimental results and flexibility in choosing the error measure make CoEvoRDT a promising approach for constructing robust DTs in real-world applications.
A plethora of fair graph neural networks (GNNs) have been proposed to promote algorithmic fairness for high-stake real-life contexts. Meanwhile, explainability is generally proposed to help machine learning practitioners debug models by providing human-understandable explanations. However, seldom work on explainability is made to generate explanations for fairness diagnosis in GNNs. From the explainability perspective, this paper explores the problem of what subgraph patterns cause the biased behavior of GNNs, and what actions could practitioners take to rectify the bias? By answering the two questions, this paper aims to produce compact, diagnostic, and actionable explanations that are responsible for discriminatory behavior. Specifically, we formulate the problem of generating diagnostic and actionable explanations as a multi-objective combinatorial optimization problem. To solve the problem, a dedicated multi-objective evolutionary algorithm is presented to ensure GNNs' explainability and fairness in one go. In particular, an influenced nodes-based gradient approximation is developed to boost the computation efficiency of the evolutionary algorithm. We provide a theoretical analysis to illustrate the effectiveness of the proposed framework. Extensive experiments have been conducted to demonstrate the superiority of the proposed method in terms of classification performance, fairness, and interpretability.
This paper reveals a data bias issue that can profoundly hinder the performance of machine learning models in malicious URL detection. We describe how such bias can be diagnosed using interpretable machine learning techniques and further argue that such biases naturally exist in the real world security data for training a classification model. To counteract these challenges, we propose a debiased training strategy that can be applied to most deep-learning based models to alleviate the negative effects of the biased features. The solution is based on the technique of adversarial training to train deep neural networks learning invariant embedding from biased data. Through extensive experimentation, we substantiate that our innovative strategy fosters superior generalization capabilities across both CNN-based and RNN-based detection models. The findings presented in this work not only expose a latent issue in the field but also provide an actionable remedy, marking a significant step forward in the pursuit of more reliable and robust malicious URL detection.
Large Language Models (LLMs) have achieved unprecedented breakthroughs in various natural language processing domains. However, the enigmatic ``black-box'' nature of LLMs remains a significant challenge for interpretability, hampering transparent and accountable applications. While past approaches, such as attention visualization, pivotal subnetwork extraction, and concept-based analyses, offer some insight, they often focus on either local or global explanations within a single dimension, occasionally falling short in providing comprehensive clarity. In response, we propose a novel methodology anchored in sparsity-guided techniques, aiming to provide a holistic interpretation of LLMs. Our framework, termed SparseCBM, innovatively integrates sparsity to elucidate three intertwined layers of interpretation: input, subnetwork, and concept levels. In addition, the newly introduced dimension of interpretable inference-time intervention facilitates dynamic adjustments to the model during deployment. Through rigorous empirical evaluations on real-world datasets, we demonstrate that SparseCBM delivers a profound understanding of LLM behaviors, setting it apart in both interpreting and ameliorating model inaccuracies. Codes are provided in supplements.
Interpretability Benchmark for Evaluating Spatial Misalignment of Prototypical Parts Explanations
PDF ↗Prototypical parts-based networks are becoming increasingly popular due to their faithful self-explanations. However, their similarity maps are calculated in the penultimate network layer. Therefore, the receptive field of the prototype activation region often depends on parts of the image outside this region, which can lead to misleading interpretations. We name this undesired behavior a spatial explanation misalignment and introduce an interpretability benchmark with a set of dedicated metrics for quantifying this phenomenon. In addition, we propose a method for misalignment compensation and apply it to existing state-of-the-art models. We show the expressiveness of our benchmark and the effectiveness of the proposed compensation methodology through extensive empirical studies.