Data labeling is often the most challenging task when developing computational pathology models. Pathologist participation is necessary to generate accurate labels, and the limitations on pathologist time and demand for large, labeled datasets has led to research in areas including weakly supervised learning using patient-level labels, machine assisted annotation and active learning. In this paper we explore self-supervised learning to reduce labeling burdens in computational pathology. We explore this in the context of classification of breast cancer tissue using the Barlow Twins approach, and we compare self-supervision with alternatives like pre-trained networks in low-data scenarios. For the task explored in this paper, we find that ImageNet pre-trained networks largely outperform the self-supervised representations obtained using Barlow Twins.
Many physics and engineering applications demand Partial Differential Equations (PDE) property evaluations that are traditionally computed with resource-intensive high-fidelity numerical solvers. Data-driven surrogate models provide an efficient alternative but come with a significant cost of training. Emerging applications would benefit from surrogates with an improved accuracy-cost tradeoff, while studied at scale. Here we present a "physics-enhanced deep-surrogate" ("PEDS") approach towards developing fast surrogate models for complex physical systems, which is described by PDEs. Specifically, a combination of a low-fidelity, explainable physics simulator and a neural network generator is proposed, which is trained end-to-end to globally match the output of an expensive high-fidelity numerical solver. Experiments on three exemplar testcases, diffusion, reaction-diffusion, and electromagnetic scattering models, show that a PEDS surrogate can be up to 3$\times$ more accurate than an ensemble of feedforward neural networks with limited data ($\approx 10^3$ training points), and reduces the training data need by at least a factor of 100 to achieve a target error of 5%. Experiments reveal that PEDS provides a general, data-driven strategy to bridge the gap between a vast array of simplified physical models with corresponding brute-force numerical solvers modeling complex systems, offering accuracy, speed, data efficiency, as well as physical insights into the process.
Despite recent progress in artificial intelligence and machine learning, many state-of-the-art methods suffer from a lack of explainability and transparency. The ability to interpret the predictions made by machine learning models and accurately evaluate these models is crucially important. In this paper, we present an interactive visualization tool to elucidate the training process of active learning. This tool enables one to select a sample of interesting data points, view how their prediction values change at different querying stages, and thus better understand when and how active learning works. Additionally, users can utilize this tool to compare different active learning strategies simultaneously and inspect why some strategies outperform others in certain contexts. With some preliminary experiments, we demonstrate that our visualization panel has a great potential to be used in various active learning experiments and help users evaluate their models appropriately.
Named entity recognition (NER) aims to identify mentions of named entities in an unstructured text and classify them into predefined named entity classes. While deep learning-based pre-trained language models help to achieve good predictive performances in NER, many domain-specific NER applications still call for a substantial amount of labeled data. Active learning (AL), a general framework for the label acquisition problem, has been used for NER tasks to minimize the annotation cost without sacrificing model performance. However, the heavily imbalanced class distribution of tokens introduces challenges in designing effective AL querying methods for NER. We propose several AL sentence query evaluation functions that pay more attention to potential positive tokens, and evaluate these proposed functions with both sentence-based and token-based cost evaluation strategies. We also propose a better data-driven normalization approach to penalize sentences that are too long or too short. Our experiments on three datasets from different domains reveal that the proposed approach reduces the number of annotated tokens while achieving better or comparable prediction performance with conventional methods.
Recently deep learning has been successfully applied to unsupervised active learning. However, current method attempts to learn a nonlinear transformation via an auto-encoder while ignoring the sample relation, leaving huge room to design more effective representation learning mechanisms for unsupervised active learning. In this paper, we propose a novel deep unsupervised Active Learning model via Learnable Graphs, named ALLG. ALLG benefits from learning optimal graph structures to acquire better sample representation and select representative samples. To make the learnt graph structure more stable and effective, we take into account $k$-nearest neighbor graph as a priori, and learn a relation propagation graph structure. We also incorporate shortcut connections among different layers, which can alleviate the well-known over-smoothing problem to some extent. To the best of our knowledge, this is the first attempt to leverage graph structure learning for unsupervised active learning. Extensive experiments performed on six datasets demonstrate the efficacy of our method.
Since GDPR came into force in May 2018, companies have worked on their data practices to comply with this privacy law. In particular, since the privacy policy is the essential communication channel for users to understand and control their privacy, many companies updated their privacy policies after GDPR was enforced. However, most privacy policies are verbose, full of jargon, and vaguely describe companies' data practices and users' rights. Therefore, it is unclear if they comply with GDPR. In this paper, we create a privacy policy dataset of 1,080 websites labeled with the 18 GDPR requirements and develop a Convolutional Neural Network (CNN) based model which can classify the privacy policies with an accuracy of 89.2%. We apply our model to perform a measurement on the compliance in the privacy policies. Our results show that even after GDPR went into effect, 97% of websites still fail to comply with at least one requirement of GDPR.
Bayesian optimization (BO) is an efficient method to optimize expensive black-box functions. It has been generalized to scenarios where objective function evaluations return stochastic binary feedback, such as success/failure in a given test, or preference between different parameter settings. In many real-world situations, the objective function can be evaluated in controlled 'contexts' or 'environments' that directly influence the observations. For example, one could directly alter the 'difficulty' of the test that is used to evaluate a system's performance. With binary feedback, the context determines the information obtained from each observation. For example, if the test is too easy/hard, the system will always succeed/fail, yielding uninformative binary outputs. Here we combine ideas from Bayesian active learning and optimization to efficiently choose the best context and optimization parameter on each iteration. We demonstrate the performance of our algorithm and illustrate how it can be used to tackle a concrete application in visual psychophysics: efficiently improving patients' vision via corrective lenses, using psychophysics measurements.
Estimating personalized treatment effects from high-dimensional observational data is essential in situations where experimental designs are infeasible, unethical, or expensive. Existing approaches rely on fitting deep models on outcomes observed for treated and control populations. However, when measuring individual outcomes is costly, as is the case of a tumor biopsy, a sample-efficient strategy for acquiring each result is required. Deep Bayesian active learning provides a framework for efficient data acquisition by selecting points with high uncertainty. However, existing methods bias training data acquisition towards regions of non-overlapping support between the treated and control populations. These are not sample-efficient because the treatment effect is not identifiable in such regions. We introduce causal, Bayesian acquisition functions grounded in information theory that bias data acquisition towards regions with overlapping support to maximize sample efficiency for learning personalized treatment effects. We demonstrate the performance of the proposed acquisition strategies on synthetic and semi-synthetic datasets IHDP and CMNIST and their extensions, which aim to simulate common dataset biases and pathologies.
The goal of a typical adaptive sequential decision making problem is to design an interactive policy that selects a group of items sequentially, based on some partial observations, to maximize the expected utility. It has been shown that the utility functions of many real-world applications, including pooled-based active learning and adaptive influence maximization, satisfy the property of adaptive submodularity. However, most of existing studies on adaptive submodular maximization focus on the fully adaptive setting, i.e., one must wait for the feedback from \emph{all} past selections before making the next selection. Although this approach can take full advantage of feedback from the past to make informed decisions, it may take a longer time to complete the selection process as compared with the non-adaptive solution where all selections are made in advance before any observations take place. In this paper, we explore the problem of partial-adaptive submodular maximization where one is allowed to make multiple selections in a batch simultaneously and observe their realizations together. Our approach enjoys the benefits of adaptivity while reducing the time spent on waiting for the observations from past selections. To the best of our knowledge, no results are known for partial-adaptive policies for the non-monotone adaptive submodular maximization problem. We study this problem under both cardinality constraint and knapsack constraints, and develop effective and efficient solutions for both cases. We also analyze the batch query complexity, i.e., the number of batches a policy takes to complete the selection process, of our policy under some additional assumptions.
One important application of transcranial magnetic stimulation (TMS) is to map cortical motor topography by spatially sampling the motor cortex, and recording motor evoked potentials (MEP) with surface electromyography. Standard approaches to TMS mapping involve repetitive stimulations at different loci spaced on a (typically 1 cm) grid on the scalp. These mappings strategies are time consuming and responsive sites are typically sparse. Furthermore, the long time scale prevents measurement of transient cortical changes, and is poorly tolerated in clinical populations. An alternative approach involves using the TMS mapper expertise to exploit the map's sparsity through the use of feedback of MEPs to decide which loci to stimulate. In this investigation, we propose a novel active learning method to automatically infer optimal future stimulus loci in place of user expertise. Specifically, we propose an active Gaussian Process (GP) strategy with loci selection criteria such as entropy and mutual information (MI). The proposed method twists the usual entropy- and MI-based selection criteria by modeling the estimated MEP field, i.e., the GP mean, as a Gaussian random variable itself. By doing so, we include MEP amplitudes in the loci selection criteria which would be otherwise completely independent of the MEP values. Experimental results using real data shows that the proposed strategy can greatly outperform competing methods when the MEP variations are mostly confined in a sub-region of the space.
The re-emergence of mosquito-borne diseases (MBDs), which kill hundreds of thousands of people each year, has been attributed to increased human population, migration, and environmental changes. Convolutional neural networks (CNNs) have been used by several studies to recognise mosquitoes in images provided by projects such as Mosquito Alert to assist entomologists in identifying, monitoring, and managing MBD. Nonetheless, utilising CNNs to automatically label input samples could involve incorrect predictions, which may mislead future epidemiological studies. Furthermore, CNNs require large numbers of manually annotated data. In order to address the mentioned issues, this paper proposes using the Monte Carlo Dropout method to estimate the uncertainty scores in order to rank the classified samples to reduce the need for human supervision in recognising Aedes albopictus mosquitoes. The estimated uncertainty was also used in an active learning framework, where just a portion of the data from large training sets was manually labelled. The experimental results show that the proposed classification method with rejection outperforms the competing methods by improving overall performance and reducing entomologist annotation workload. We also provide explainable visualisations of the different regions that contribute to a set of samples' uncertainty assessment.
Uncertainty sampling in active learning is heavily used in practice to reduce the annotation cost. However, there has been no wide consensus on the function to be used for uncertainty estimation in binary classification tasks and convergence guarantees of the corresponding active learning algorithms are not well understood. The situation is even more challenging for multi-category classification. In this work, we propose an efficient uncertainty estimator for binary classification which we also extend to multiple classes, and provide a non-asymptotic rate of convergence for our uncertainty sampling-based active learning algorithm in both cases under no-noise conditions (i.e., linearly separable data). We also extend our analysis to the noisy case and provide theoretical guarantees for our algorithm under the influence of noise in the task of binary and multi-class classification.
We study the problem of active learning with the added twist that the learner is assisted by a helpful teacher. We consider the following natural interaction protocol: At each round, the learner proposes a query asking for the label of an instance $x^q$, the teacher provides the requested label $\{x^q, y^q\}$ along with explanatory information to guide the learning process. In this paper, we view this information in the form of an additional contrastive example ($\{x^c, y^c\}$) where $x^c$ is picked from a set constrained by $x^q$ (e.g., dissimilar instances with the same label). Our focus is to design a teaching algorithm that can provide an informative sequence of contrastive examples to the learner to speed up the learning process. We show that this leads to a challenging sequence optimization problem where the algorithm's choices at a given round depend on the history of interactions. We investigate an efficient teaching algorithm that adaptively picks these contrastive examples. We derive strong performance guarantees for our algorithm based on two problem-dependent parameters and further show that for specific types of active learners (e.g., a generalized binary search learner), the proposed teaching algorithm exhibits strong approximation guarantees. Finally, we illustrate our bounds and demonstrate the effectiveness of our teaching framework via two numerical case studies.
Message passing is the core of most graph models such as Graph Convolutional Network (GCN) and Label Propagation (LP), which usually require a large number of clean labeled data to smooth out the neighborhood over the graph. However, the labeling process can be tedious, costly, and error-prone in practice. In this paper, we propose to unify active learning (AL) and message passing towards minimizing labeling costs, e.g., making use of few and unreliable labels that can be obtained cheaply. We make two contributions towards that end. First, we open up a perspective by drawing a connection between AL enforcing message passing and social influence maximization, ensuring that the selected samples effectively improve the model performance. Second, we propose an extension to the influence model that incorporates an explicit quality factor to model label noise. In this way, we derive a fundamentally new AL selection criterion for GCN and LP--reliable influence maximization (RIM)--by considering quantity and quality of influence simultaneously. Empirical studies on public datasets show that RIM significantly outperforms current AL methods in terms of accuracy and efficiency.
With a principled representation of uncertainty and closed form posterior updates, Gaussian processes (GPs) are a natural choice for online decision making. However, Gaussian processes typically require at least $\mathcal{O}(n^2)$ computations for $n$ training points, limiting their general applicability. Stochastic variational Gaussian processes (SVGPs) can provide scalable inference for a dataset of fixed size, but are difficult to efficiently condition on new data. We propose online variational conditioning (OVC), a procedure for efficiently conditioning SVGPs in an online setting that does not require re-training through the evidence lower bound with the addition of new data. OVC enables the pairing of SVGPs with advanced look-ahead acquisition functions for black-box optimization, even with non-Gaussian likelihoods. We show OVC provides compelling performance in a range of applications including active learning of malaria incidence, and reinforcement learning on MuJoCo simulated robotic control tasks.
Active learning is a subfield of machine learning that is devised for design and modeling of systems with highly expensive sampling costs. Industrial and engineering systems are generally subject to physics constraints that may induce fatal failures when they are violated, while such constraints are frequently underestimated in active learning. In this paper, we develop a novel active learning method that avoids failures considering implicit physics constraints that govern the system. The proposed approach is driven by two tasks: the safe variance reduction explores the safe region to reduce the variance of the target model, and the safe region expansion aims to extend the explorable region exploiting the probabilistic model of constraints. The global acquisition function is devised to judiciously optimize acquisition functions of two tasks, and its theoretical properties are provided. The proposed method is applied to the composite fuselage assembly process with consideration of material failure using the Tsai-wu criterion, and it is able to achieve zero-failure without the knowledge of explicit failure regions.
The Chow-Liu algorithm (IEEE Trans.~Inform.~Theory, 1968) has been a mainstay for the learning of tree-structured graphical models from i.i.d.\ sampled data vectors. Its theoretical properties have been well-studied and are well-understood. In this paper, we focus on the class of trees that are arguably even more fundamental, namely {\em homogeneous} trees in which each pair of nodes that forms an edge has the same correlation $\rho$. We ask whether we are able to further reduce the error probability of learning the structure of the homogeneous tree model when {\em active learning} or {\em active sampling of nodes or variables} is allowed. Our figure of merit is the {\em error exponent}, which quantifies the exponential rate of decay of the error probability with an increasing number of data samples. At first sight, an improvement in the error exponent seems impossible, as all the edges are statistically identical. We design and analyze an algorithm Active Learning Algorithm for Trees with Homogeneous Edge (Active-LATHE), which surprisingly boosts the error exponent by at least 40\% when $\rho$ is at least $0.8$. For all other values of $\rho$, we also observe commensurate, but more modest, improvements in the error exponent. Our analysis hinges on judiciously exploiting the minute but detectable statistical variation of the samples to allocate more data to parts of the graph in which we are less confident of being correct.
We study acquisition functions for active learning (AL) for text classification. The Expected Loss Reduction (ELR) method focuses on a Bayesian estimate of the reduction in classification error, recently updated with Mean Objective Cost of Uncertainty (MOCU). We convert the ELR framework to estimate the increase in (strictly proper) scores like log probability or negative mean square error, which we call Bayesian Estimate of Mean Proper Scores (BEMPS). We also prove convergence results borrowing techniques used with MOCU. In order to allow better experimentation with the new acquisition functions, we develop a complementary batch AL algorithm, which encourages diversity in the vector of expected changes in scores for unlabelled data. To allow high performance text classifiers, we combine ensembling and dynamic validation set construction on pretrained language models. Extensive experimental evaluation then explores how these different acquisition functions perform. The results show that the use of mean square error and log probability with BEMPS yields robust acquisition functions, which consistently outperform the others tested.
Convolutional neural networks (CNNs) have been successfully applied to the single target tracking task in recent years. Generally, training a deep CNN model requires numerous labeled training samples, and the number and quality of these samples directly affect the representational capability of the trained model. However, this approach is restrictive in practice, because manually labeling such a large number of training samples is time-consuming and prohibitively expensive. In this paper, we propose an active learning method for deep visual tracking, which selects and annotates the unlabeled samples to train the deep CNNs model. Under the guidance of active learning, the tracker based on the trained deep CNNs model can achieve competitive tracking performance while reducing the labeling cost. More specifically, to ensure the diversity of selected samples, we propose an active learning method based on multi-frame collaboration to select those training samples that should be and need to be annotated. Meanwhile, considering the representativeness of these selected samples, we adopt a nearest neighbor discrimination method based on the average nearest neighbor distance to screen isolated samples and low-quality samples. Therefore, the training samples subset selected based on our method requires only a given budget to maintain the diversity and representativeness of the entire sample set. Furthermore, we adopt a Tversky loss to improve the bounding box estimation of our tracker, which can ensure that the tracker achieves more accurate target states. Extensive experimental results confirm that our active learning-based tracker (ALT) achieves competitive tracking accuracy and speed compared with state-of-the-art trackers on the seven most challenging evaluation benchmarks.
Online relevance feedback (RF) is widely utilized in instance search (INS) tasks to further refine imperfect ranking results, but it often has low interaction efficiency. The active learning (AL) technique addresses this problem by selecting valuable feedback candidates. However, mainstream AL methods require an initial labeled set for a cold start and are often computationally complex to solve. Therefore, they cannot fully satisfy the requirements for online RF in interactive INS tasks. To address this issue, we propose a confidence-aware active feedback method (CAAF) that is specifically designed for online RF in interactive INS tasks. Inspired by the explicit difficulty modeling scheme in self-paced learning, CAAF utilizes a pairwise manifold ranking loss to evaluate the ranking confidence of each unlabeled sample. The ranking confidence improves not only the interaction efficiency by indicating valuable feedback candidates but also the ranking quality by modulating the diffusion weights in manifold ranking. In addition, we design two acceleration strategies, an approximate optimization scheme and a top-K search scheme, to reduce the computational complexity of CAAF. Extensive experiments on both image INS tasks and video INS tasks searching for buildings, landscapes, persons, and human behaviors demonstrate the effectiveness of the proposed method. Notably, in the real-world, large-scale video INS task of NIST TRECVID 2021, CAAF uses 25% fewer feedback samples to achieve a performance that is nearly equivalent to the champion solution. Moreover, with the same number of feedback samples, CAAF's mAP is 51.9%, significantly surpassing the champion solution by 5.9%. Code is available at https://github.com/nercms-mmap/caaf.