Class Incremental Learning (CIL) based on pre-trained models offers a promising direction for open-world continual learning. Existing methods typically rely on correlation-based strategies, where an image's classification feature is used as a query to retrieve the most related key prompts and select the corresponding value prompts for training. However, these approaches face an inherent limitation: fitting the entire feature space of all tasks with only a few trainable prompts is fundamentally challenging. We propose Predictive Prompting (PrePrompt), a novel CIL framework that circumvents correlation-based limitations by leveraging pre-trained models' natural classification ability to predict task-specific prompts. Specifically, PrePrompt decomposes CIL into a two-stage prediction framework: task-specific prompt prediction followed by label prediction. While theoretically appealing, this framework risks bias toward recent classes due to missing historical data for older classifier calibration. PrePrompt then mitigates this by incorporating feature translation, dynamically balancing stability and plasticity. Experiments across multiple benchmarks demonstrate PrePrompt's superiority over state-of-the-art prompt-based CIL methods. Code available at \href{github.com/libo-huang/preprompt}{github.com/libo-huang/preprompt}.
Recently computer-aided diagnosis has demonstrated promising performance, effectively alleviating the workload of clinicians. However, the inherent sample imbalance among different diseases leads algorithms biased to the majority categories, leading to poor performance for rare categories. Existing works formulated this challenge as a long-tailed problem and attempted to tackle it by decoupling the feature representation and classification. Yet, due to the imbalanced distribution and limited samples from tail classes, these works are prone to biased representation learning and insufficient classifier calibration. To tackle these problems, we propose a new Long-tailed Medical Diagnosis (LMD) framework for balanced medical image classification on long-tailed datasets. In the initial stage, we develop a Relation-aware Representation Learning (RRL) scheme to boost the representation ability by encouraging the encoder to capture intrinsic semantic features through different data augmentations. In the subsequent stage, we propose an Iterative Classifier Calibration (ICC) scheme to calibrate the classifier iteratively. This is achieved by generating a large number of balanced virtual features and fine-tuning the encoder using an Expectation-Maximization manner. The proposed ICC compensates for minority categories to facilitate unbiased classifier optimization while maintaining the diagnostic knowledge in majority classes. Comprehensive experiments on three public long-tailed medical datasets demonstrate that our LMD framework significantly surpasses state-of-the-art approaches. The source code can be accessed at https://github.com/peterlipan/LMD.
Prototype-based federated learning has emerged as a promising approach that shares lightweight prototypes to transfer knowledge among clients with data heterogeneity in a model-agnostic manner. However, existing methods often collect prototypes directly from local models, which inevitably introduce inconsistencies into representation learning due to the biased data distributions and differing model architectures among clients. In this paper, we identify that both statistical and model heterogeneity create a vicious cycle of representation inconsistency, classifier divergence, and skewed prototype alignment, which negatively impacts the performance of clients. To break the vicious cycle, we propose a novel framework named Federated Learning via Semantic Anchors (FedSA) to decouple the generation of prototypes from local representation learning. We introduce a novel perspective that uses simple yet effective semantic anchors serving as prototypes to guide local models in learning consistent representations. By incorporating semantic anchors, we further propose anchor-based regularization with margin-enhanced contrastive learning and anchor-based classifier calibration to correct feature extractors and calibrate classifiers across clients, achieving intra-class compactness and inter-class separability of prototypes while ensuring consistent decision boundaries. We then update the semantic anchors with these consistent and discriminative prototypes, which iteratively encourage clients to collaboratively learn a unified data representation with robust generalization. Extensive experiments under both statistical and model heterogeneity settings show that FedSA significantly outperforms existing prototype-based FL methods on various classification tasks.
Generalized few-shot semantic segmentation (GFSS) aims to segment objects of both base and novel classes, using sufficient samples of base classes and few samples of novel classes. Representative GFSS approaches typically employ a two-phase training scheme, involving base class pre-training followed by novel class fine-tuning, to learn the classifiers for base and novel classes respectively. Nevertheless, distribution gap exists between base and novel classes in this process. To narrow this gap, we exploit effective knowledge transfer from base to novel classes. First, a novel prototype modulation module is designed to modulate novel class prototypes by exploiting the correlations between base and novel classes. Second, a novel classifier calibration module is proposed to calibrate the weight distribution of the novel classifier according to that of the base classifier. Furthermore, existing GFSS approaches suffer from a lack of contextual information for novel classes due to their limited samples, we thereby introduce a context consistency learning scheme to transfer the contextual knowledge from base to novel classes. Extensive experiments on PASCAL-5$^i$ and COCO-20$^i$ demonstrate that our approach significantly enhances the state of the art in the GFSS setting. The code is available at: https://github.com/HHHHedy/GFSS-EKT.
Classification models based on deep neural networks (DNNs) must be calibrated to measure the reliability of predictions. Some recent calibration methods have employed a probabilistic model on the probability simplex. However, these calibration methods cannot preserve the accuracy of pre-trained models, even those with a high classification accuracy. We propose an accuracy-preserving calibration method using the Concrete distribution as the probabilistic model on the probability simplex. We theoretically prove that a DNN model trained on cross-entropy loss has optimality as the parameter of the Concrete distribution. We also propose an efficient method that synthetically generates samples for training probabilistic models on the probability simplex. We demonstrate that the proposed method can outperform previous methods in accuracy-preserving calibration tasks using benchmarks. The code is available at https://github.com/ToyotaCRDL/SimplexTS.
Machine learning models typically focus on specific targets like creating classifiers, often based on known population feature distributions in a business context. However, models calculating individual features adapt over time to improve precision, introducing the concept of decoupling: shifting from point evaluation to data distribution. We use calibration strategies as strategy for decoupling machine learning (ML) classifiers from score-based actions within business logic frameworks. To evaluate these strategies, we perform a comparative analysis using a real-world business scenario and multiple ML models. Our findings highlight the trade-offs and performance implications of the approach, offering valuable insights for practitioners seeking to optimize their decoupling efforts. In particular, the Isotonic and Beta calibration methods stand out for scenarios in which there is shift between training and testing data.
Calibration of machine learning classifiers is necessary to obtain reliable and interpretable predictions, bridging the gap between model confidence and actual probabilities. One prominent technique, isotonic regression (IR), aims at calibrating binary classifiers by minimizing the cross entropy on a calibration set via monotone transformations. IR acts as an adaptive binning procedure, which allows achieving a calibration error of zero, but leaves open the issue of the effect on performance. In this paper, we first prove that IR preserves the convex hull of the ROC curve -- an essential performance metric for binary classifiers. This ensures that a classifier is calibrated while controlling for overfitting of the calibration set. We then present a novel generalization of isotonic regression to accommodate classifiers with K classes. Our method constructs a multidimensional adaptive binning scheme on the probability simplex, again achieving a multi-class calibration error equal to zero. We regularize this algorithm by imposing a form of monotony that preserves the K-dimensional ROC surface of the classifier. We show empirically that this general monotony criterion is effective in striking a balance between reducing cross entropy loss and avoiding overfitting of the calibration set.
Data heterogeneity is an inherent challenge that hinders the performance of federated learning (FL). Recent studies have identified the biased classifiers of local models as the key bottleneck. Previous attempts have used classifier calibration after FL training, but this approach falls short in improving the poor feature representations caused by training-time classifier biases. Resolving the classifier bias dilemma in FL requires a full understanding of the mechanisms behind the classifier. Recent advances in neural collapse have shown that the classifiers and feature prototypes under perfect training scenarios collapse into an optimal structure called simplex equiangular tight frame (ETF). Building on this neural collapse insight, we propose a solution to the FL's classifier bias problem by utilizing a synthetic and fixed ETF classifier during training. The optimal classifier structure enables all clients to learn unified and optimal feature representations even under extremely heterogeneous data. We devise several effective modules to better adapt the ETF structure in FL, achieving both high generalization and personalization. Extensive experiments demonstrate that our method achieves state-of-the-art performances on CIFAR-10, CIFAR-100, and Tiny-ImageNet.
Mix-up training approaches have proven to be effective in improving the generalization ability of Deep Neural Networks. Over the years, the research community expands mix-up methods into two directions, with extensive efforts to improve saliency-guided procedures but minimal focus on the arbitrary path, leaving the randomization domain unexplored. In this paper, inspired by the superior qualities of each direction over one another, we introduce a novel method that lies at the junction of the two routes. By combining the best elements of randomness and saliency utilization, our method balances speed, simplicity, and accuracy. We name our method R-Mix following the concept of "Random Mix-up". We demonstrate its effectiveness in generalization, weakly supervised object localization, calibration, and robustness to adversarial attacks. Finally, in order to address the question of whether there exists a better decision protocol, we train a Reinforcement Learning agent that decides the mix-up policies based on the classifier's performance, reducing dependency on human-designed objectives and hyperparameter tuning. Extensive experiments further show that the agent is capable of performing at the cutting-edge level, laying the foundation for a fully automatic mix-up. Our code is released at [https://github.com/minhlong94/Random-Mixup].
Federated learning allows multiple clients to collaboratively train a model without exchanging their data, thus preserving data privacy. Unfortunately, it suffers significant performance degradation due to heterogeneous data at clients. Common solutions involve designing an auxiliary loss to regularize weight divergence or feature inconsistency during local training. However, we discover that these approaches fall short of the expected performance because they ignore the existence of a vicious cycle between feature inconsistency and classifier divergence across clients. This vicious cycle causes client models to be updated in inconsistent feature spaces with more diverged classifiers. To break the vicious cycle, we propose a novel framework named Federated learning with Feature Anchors (FedFA). FedFA utilizes feature anchors to align features and calibrate classifiers across clients simultaneously. This enables client models to be updated in a shared feature space with consistent classifiers during local training. Theoretically, we analyze the non-convex convergence rate of FedFA. We also demonstrate that the integration of feature alignment and classifier calibration in FedFA brings a virtuous cycle between feature and classifier updates, which breaks the vicious cycle existing in current approaches. Extensive experiments show that FedFA significantly outperforms existing approaches on various classification datasets under label distribution skew and feature distribution skew.
Deep Ensembles (DE) are a prominent approach for achieving excellent performance on key metrics such as accuracy, calibration, uncertainty estimation, and out-of-distribution detection. However, hardware limitations of real-world systems constrain to smaller ensembles and lower-capacity networks, significantly deteriorating their performance and properties. We introduce Packed-Ensembles (PE), a strategy to design and train lightweight structured ensembles by carefully modulating the dimension of their encoding space. We leverage grouped convolutions to parallelize the ensemble into a single shared backbone and forward pass to improve training and inference speeds. PE is designed to operate within the memory limits of a standard neural network. Our extensive research indicates that PE accurately preserves the properties of DE, such as diversity, and performs equally well in terms of accuracy, calibration, out-of-distribution detection, and robustness to distribution shift. We make our code available at https://github.com/ENSTA-U2IS/torch-uncertainty.
For many applications of probabilistic classifiers it is important that the predicted confidence vectors reflect true probabilities (one says that the classifier is calibrated). It has been shown that common models fail to satisfy this property, making reliable methods for measuring and improving calibration important tools. Unfortunately, obtaining these is far from trivial for problems with many classes. We propose two techniques that can be used in tandem. First, a reduced calibration method transforms the original problem into a simpler one. We prove for several notions of calibration that solving the reduced problem minimizes the corresponding notion of miscalibration in the full problem, allowing the use of non-parametric recalibration methods that fail in higher dimensions. Second, we propose class-wise calibration methods, based on intuition building on a phenomenon called neural collapse and the observation that most of the accurate classifiers found in practice can be thought of as a union of K different functions which can be recalibrated separately, one for each class. These typically out-perform their non class-wise counterparts, especially for classifiers trained on imbalanced data sets. Applying the two methods together results in class-wise reduced calibration algorithms, which are powerful tools for reducing the prediction and per-class calibration errors. We demonstrate our methods on real and synthetic datasets and release all code as open source at https://github.com/appliedAI-Initiative
Classifier calibration has received recent attention from the machine learning community due both to its practical utility in facilitating decision making, as well as the observation that modern neural network classifiers are poorly calibrated. Much of this focus has been towards the goal of learning classifiers such that their output with largest magnitude (the "predicted class") is calibrated. However, this narrow interpretation of classifier outputs does not adequately capture the variety of practical use cases in which classifiers can aid in decision making. In this work, we argue that more expressive metrics must be developed that accurately measure calibration error for the specific context in which a classifier will be deployed. To this end, we derive a number of different metrics using a generalization of Expected Calibration Error (ECE) that measure calibration error under different definitions of reliability. We then provide an extensive empirical evaluation of commonly used neural network architectures and calibration techniques with respect to these metrics. We find that: 1) definitions of ECE that focus solely on the predicted class fail to accurately measure calibration error under a selection of practically useful definitions of reliability and 2) many common calibration techniques fail to improve calibration performance uniformly across ECE metrics derived from these diverse definitions of reliability.
Trustworthy classifiers are essential to the adoption of machine learning predictions in many real-world settings. The predicted probability of possible outcomes can inform high-stakes decision making, particularly when assessing the expected value of alternative decisions or the risk of bad outcomes. These decisions require well-calibrated probabilities, not just the correct prediction of the most likely class. Black-box classifier calibration methods can improve the reliability of a classifier's output without requiring retraining. However, these methods are unable to detect subpopulations where calibration could also improve prediction accuracy. Such subpopulations are said to exhibit "hidden heterogeneity" (HH), because the original classifier did not detect them. This paper proposes a quantitative measure for HH. It also introduces two similarity-weighted calibration methods that can address HH by adapting locally to each test item: SWC weights the calibration set by similarity to the test item, and SWC-HH explicitly incorporates hidden heterogeneity to filter the calibration set. Experiments show that the improvements in calibration achieved by similarity-based calibration methods correlate with the amount of HH present and, given sufficient calibration data, generally exceed calibration achieved by global methods. HH can therefore serve as a useful diagnostic tool for identifying when local calibration methods would be beneficial.
This paper provides both an introduction to and a detailed overview of the principles and practice of classifier calibration. A well-calibrated classifier correctly quantifies the level of uncertainty or confidence associated with its instance-wise predictions. This is essential for critical applications, optimal decision making, cost-sensitive classification, and for some types of context change. Calibration research has a rich history which predates the birth of machine learning as an academic field by decades. However, a recent increase in the interest on calibration has led to new methods and the extension from binary to the multiclass setting. The space of options and issues to consider is large, and navigating it requires the right set of concepts and tools. We provide both introductory material and up-to-date technical details of the main concepts and methods, including proper scoring rules and other evaluation metrics, visualisation approaches, a comprehensive account of post-hoc calibration methods for binary and multiclass classification, and several advanced topics.
Both generalized and incremental few-shot learning have to deal with three major challenges: learning novel classes from only few samples per class, preventing catastrophic forgetting of base classes, and classifier calibration across novel and base classes. In this work we propose a three-stage framework that allows to explicitly and effectively address these challenges. While the first phase learns base classes with many samples, the second phase learns a calibrated classifier for novel classes from few samples while also preventing catastrophic forgetting. In the final phase, calibration is achieved across all classes. We evaluate the proposed framework on four challenging benchmark datasets for image and video few-shot classification and obtain state-of-the-art results for both generalized and incremental few shot learning.
A central challenge in training classification models in the real-world federated system is learning with non-IID data. To cope with this, most of the existing works involve enforcing regularization in local optimization or improving the model aggregation scheme at the server. Other works also share public datasets or synthesized samples to supplement the training of under-represented classes or introduce a certain level of personalization. Though effective, they lack a deep understanding of how the data heterogeneity affects each layer of a deep classification model. In this paper, we bridge this gap by performing an experimental analysis of the representations learned by different layers. Our observations are surprising: (1) there exists a greater bias in the classifier than other layers, and (2) the classification performance can be significantly improved by post-calibrating the classifier after federated training. Motivated by the above findings, we propose a novel and simple algorithm called Classifier Calibration with Virtual Representations (CCVR), which adjusts the classifier using virtual representations sampled from an approximated gaussian mixture model. Experimental results demonstrate that CCVR achieves state-of-the-art performance on popular federated learning benchmarks including CIFAR-10, CIFAR-100, and CINIC-10. We hope that our simple yet effective method can shed some light on the future research of federated learning with non-IID data.
We introduce a novel fine-grained dataset and benchmark, the Danish Fungi 2020 (DF20). The dataset, constructed from observations submitted to the Atlas of Danish Fungi, is unique in its taxonomy-accurate class labels, small number of errors, highly unbalanced long-tailed class distribution, rich observation metadata, and well-defined class hierarchy. DF20 has zero overlap with ImageNet, allowing unbiased comparison of models fine-tuned from publicly available ImageNet checkpoints. The proposed evaluation protocol enables testing the ability to improve classification using metadata -- e.g. precise geographic location, habitat, and substrate, facilitates classifier calibration testing, and finally allows to study the impact of the device settings on the classification performance. Experiments using Convolutional Neural Networks (CNN) and the recent Vision Transformers (ViT) show that DF20 presents a challenging task. Interestingly, ViT achieves results superior to CNN baselines with 80.45% accuracy and 0.743 macro F1 score, reducing the CNN error by 9% and 12% respectively. A simple procedure for including metadata into the decision process improves the classification accuracy by more than 2.95 percentage points, reducing the error rate by 15%. The source code for all methods and experiments is available at https://sites.google.com/view/danish-fungi-dataset.
This paper explores the calibration of a classifier output score in binary classification problems. A calibrator is a function that maps the arbitrary classifier score, of a testing observation, onto $[0,1]$ to provide an estimate for the posterior probability of belonging to one of the two classes. Calibration is important for two reasons; first, it provides a meaningful score, that is the posterior probability; second, it puts the scores of different classifiers on the same scale for comparable interpretation. The paper presents three main contributions: (1) Introducing multi-score calibration, when more than one classifier provides a score for a single observation. (2) Introducing the idea that the classifier scores to a calibration process are nothing but features to a classifier, hence proposing expanding the classifier scores to higher dimensions to boost the calibrator's performance. (3) Conducting a massive simulation study, in the order of 24,000 experiments, that incorporates different configurations, in addition to experimenting on two real datasets from the cybersecurity domain. The results show that there is no overall winner among the different calibrators and different configurations. However, general advices for practitioners include the following: the Platt's calibrator~\citep{Platt1999ProbabilisticOutputsForSupport}, a version of the logistic regression that decreases bias for a small sample size, has a very stable and acceptable performance among all experiments; our suggested multi-score calibration provides better performance than single score calibration in the majority of experiments, including the two real datasets. In addition, expanding the scores can help in some experiments.
Deep neural networks have amply demonstrated their prowess but estimating the reliability of their predictions remains challenging. Deep Ensembles are widely considered as being one of the best methods for generating uncertainty estimates but are very expensive to train and evaluate. MC-Dropout is another popular alternative, which is less expensive, but also less reliable. Our central intuition is that there is a continuous spectrum of ensemble-like models of which MC-Dropout and Deep Ensembles are extreme examples. The first uses an effectively infinite number of highly correlated models while the second relies on a finite number of independent models. To combine the benefits of both, we introduce Masksembles. Instead of randomly dropping parts of the network as in MC-dropout, Masksemble relies on a fixed number of binary masks, which are parameterized in a way that allows to change correlations between individual models. Namely, by controlling the overlap between the masks and their density one can choose the optimal configuration for the task at hand. This leads to a simple and easy to implement method with performance on par with Ensembles at a fraction of the cost. We experimentally validate Masksembles on two widely used datasets, CIFAR10 and ImageNet.