skip to main content


Title: Active Bayesian Assessment of Black-Box Classifiers
Recent advances in machine learning have led to increased deployment of black-box classifiers across a wide variety of applications. In many such situations there is a critical need to both reliably assess the performance of these pre-trained models and to perform this assessment in a label-efficient manner (given that labels may be scarce and costly to collect). In this paper, we introduce an active Bayesian approach for assessment of classifier performance to satisfy the desiderata of both reliability and label-efficiency. We begin by developing inference strategies to quantify uncertainty for common assessment metrics such as accuracy, misclassification cost, and calibration error. We then propose a general framework for active Bayesian assessment using inferred uncertainty to guide efficient selection of instances for labeling, enabling better performance assessment with fewer labels. We demonstrate significant gains from our proposed active Bayesian approach via a series of systematic empirical experiments assessing the performance of modern neural classifiers (e.g., ResNet and BERT) on several standard image and text classification datasets.  more » « less
Award ID(s):
1927245 1900644
NSF-PAR ID:
10280592
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
Proceedings of the AAAI Conference on Artificial Intelligence
Volume:
35
Issue:
9
ISSN:
2374-3468
Page Range / eLocation ID:
7935-7944
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Active learning methods, like uncertainty sampling, combined with probabilistic prediction techniques have achieved success in various problems like image classification and text classification. For more complex multivariate prediction tasks, the relationships between labels play an important role in designing structured classifiers with better performance. However, computational time complexity limits prevalent probabilistic methods from effectively supporting active learning. Specifically, while non-probabilistic methods based on structured support vector ma-chines can be tractably applied to predicting cuts and bipartite matchings, conditional random fields are intractable for these structures. We propose an adversarial approach for active learning with structured prediction domains that is tractable for cuts and matching. We evaluate this approach algorithmically in two important structured prediction problems: multi-label classification and object tracking in videos. We demonstrate better accuracy and computational efficiency for our proposed method. 
    more » « less
  2. How can we collect the most useful labels to learn a model selection policy, when presented with arbitrary heterogeneous data streams? In this paper, we formulate this task as a contextual active model selection problem, where at each round the learner receives an unlabeled data point along with a context. The goal is to output the best model for any given context without obtaining an excessive amount of labels. In particular, we focus on the task of selecting pre-trained classifiers, and propose a contextual active model selection algorithm (CAMS), which relies on a novel uncertainty sampling query criterion defined on a given policy class for adaptive model selection. In comparison to prior art, our algorithm does not assume a globally optimal model. We provide rigorous theoretical analysis for the regret and query complexity under both adversarial and stochastic settings. Our experiments on several benchmark classification datasets demonstrate the algorithm’s effectiveness in terms of both regret and query complexity. Notably, to achieve the same accuracy, CAMS incurs less than 10% of the label cost when compared to the best online model selection baselines on CIFAR10. 
    more » « less
  3. null (Ed.)
    Abstract Neural state classification (NSC) is a recently proposed method for runtime predictive monitoring of hybrid automata (HA) using deep neural networks (DNNs). NSC trains a DNN as an approximate reachability predictor that labels an HA state x as positive if an unsafe state is reachable from x within a given time bound, and labels x as negative otherwise. NSC predictors have very high accuracy, yet are prone to prediction errors that can negatively impact reliability. To overcome this limitation, we present neural predictive monitoring (NPM), a technique that complements NSC predictions with estimates of the predictive uncertainty. These measures yield principled criteria for the rejection of predictions likely to be incorrect, without knowing the true reachability values. We also present an active learning method that significantly reduces the NSC predictor’s error rate and the percentage of rejected predictions. We develop two versions of NPM based, respectively, on the use of frequentist and Bayesian techniques to learn the predictor and the rejection rule. Both versions are highly efficient, with computation times on the order of milliseconds, and effective, managing in our experimental evaluation to successfully reject almost all incorrect predictions. In our experiments on a benchmark suite of six hybrid systems, we found that the frequentist approach consistently outperforms the Bayesian one. We also observed that the Bayesian approach is less practical, requiring a careful and problem-specific choice of hyperparameters. 
    more » « less
  4. In label-noise learning, estimating the transition matrix is a hot topic as the matrix plays an important role in building statistically consistent classifiers. Traditionally, the transition from clean labels to noisy labels (i.e., clean-label transition matrix (CLTM)) has been widely exploited to learn a clean label classifier by employing the noisy data. Motivated by that classifiers mostly output Bayes optimal labels for prediction, in this paper, we study to directly model the transition from Bayes optimal labels to noisy labels (i.e., Bayes-label transition matrix (BLTM)) and learn a classifier to predict Bayes optimal labels. Note that given only noisy data, it is ill-posed to estimate either the CLTM or the BLTM. But favorably, Bayes optimal labels have less uncertainty compared with the clean labels, i.e., the class posteriors of Bayes optimal labels are one-hot vectors while those of clean labels are not. This enables two advantages to estimate the BLTM, i.e., (a) a set of examples with theoretically guaranteed Bayes optimal labels can be collected out of noisy data; (b) the feasible solution space is much smaller. By exploiting the advantages, we estimate the BLTM parametrically by employing a deep neural network, leading to better generalization and superior classification performance. 
    more » « less
  5. Semi-supervised learning uses underlying relationships in data with a scarcity of ground-truth labels. In this paper, we introduce an uncertainty quantification (UQ) method for graph-based semi-supervised multi-class classification problems. We not only predict the class label for each data point, but also provide a confidence score for the prediction. We adopt a Bayesian approach and propose a graphical multi-class probit model together with an effective Gibbs sampling procedure. Furthermore, we propose a confidence measure for each data point that correlates with the classification performance. We use the empirical properties of the proposed confidence measure to guide the design of a humanin-the-loop system. The uncertainty quantification algorithm and the human-in-the-loop system are successfully applied to classification problems in image processing and ego-motion analysis of body-worn videos. 
    more » « less