PyTorchKR Papers

최신 논문

총 1,324편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2304.10103

eTag: Class-Incremental Learning with Embedding Distillation and Task-Oriented Generation

Class-Incremental Learning (CIL) aims to solve the neural networks' catastrophic forgetting problem, which refers to the fact that once the network updates on a new task, its performance on previously-learned tasks drops dramatically. Most successful CIL methods incrementally train a feature extractor with the aid of stored exemplars, or estimate the feature distribution with the stored prototypes. However, the stored exemplars would violate the data privacy concerns, while the stored prototypes might not reasonably be consistent with a proper feature distribution, hindering the exploration of real-world CIL applications. In this paper, we propose a method of \textit{e}mbedding distillation and \textit{Ta}sk-oriented \textit{g}eneration (\textit{eTag}) for CIL, which requires neither the exemplar nor the prototype. Instead, eTag achieves a data-free manner to train the neural networks incrementally. To prevent the feature extractor from forgetting, eTag distills the embeddings of the network's intermediate blocks. Additionally, eTag enables a generative network to produce suitable features, fitting the needs of the top incremental classifier. Experimental results confirmed that our proposed eTag considerably outperforms the state-of-the-art methods on CIFAR-100 and ImageNet-sub\footnote{Our code is available in the Supplementary Materials.

arXiv:2304.10038

Open-World Continual Learning: Unifying Novelty Detection and Continual Learning

As AI agents are increasingly used in the real open world with unknowns or novelties, they need the ability to (1) recognize objects that (a) they have learned before and (b) detect items that they have never seen or learned, and (2) learn the new items incrementally to become more and more knowledgeable and powerful. (1) is called novelty detection or out-of-distribution (OOD) detection and (2) is called class incremental learning (CIL), which is a setting of continual learning (CL). In existing research, OOD detection and CIL are regarded as two completely different problems. This paper first provides a theoretical proof that good OOD detection for each task within the set of learned tasks (called closed-world OOD detection) is necessary for successful CIL. We show this by decomposing CIL into two sub-problems: within-task prediction (WP) and task-id prediction (TP), and proving that TP is correlated with closed-world OOD detection. The key theoretical result is that regardless of whether WP and OOD detection (or TP) are defined explicitly or implicitly by a CIL algorithm, good WP and good closed-world OOD detection are necessary and sufficient conditions for good CIL, which unifies novelty or OOD detection and continual learning (CIL, in particular). We call this traditional CIL the closed-world CIL as it does not detect future OOD data in the open world. The paper then proves that the theory can be generalized or extended to open-world CIL, which is the proposed open-world continual learning, that can perform CIL in the open world and detect future or open-world OOD data. Based on the theoretical results, new CIL methods are also designed, which outperform strong baselines in CIL accuracy and in continual OOD detection by a large margin.

arXiv:2304.09062

Always Strengthen Your Strengths: A Drift-Aware Incremental Learning Framework for CTR Prediction

Click-through rate (CTR) prediction is of great importance in recommendation systems and online advertising platforms. When served in industrial scenarios, the user-generated data observed by the CTR model typically arrives as a stream. Streaming data has the characteristic that the underlying distribution drifts over time and may recur. This can lead to catastrophic forgetting if the model simply adapts to new data distribution all the time. Also, it's inefficient to relearn distribution that has been occurred. Due to memory constraints and diversity of data distributions in large-scale industrial applications, conventional strategies for catastrophic forgetting such as replay, parameter isolation, and knowledge distillation are difficult to be deployed. In this work, we design a novel drift-aware incremental learning framework based on ensemble learning to address catastrophic forgetting in CTR prediction. With explicit error-based drift detection on streaming data, the framework further strengthens well-adapted ensembles and freezes ensembles that do not match the input distribution avoiding catastrophic interference. Both evaluations on offline experiments and A/B test shows that our method outperforms all baselines considered.

arXiv:2304.07707

Non-exemplar Class-incremental Learning by Random Auxiliary Classes Augmentation and Mixed Features

Non-exemplar class-incremental learning refers to classifying new and old classes without storing samples of old classes. Since only new class samples are available for optimization, it often occurs catastrophic forgetting of old knowledge. To alleviate this problem, many new methods are proposed such as model distillation, class augmentation. In this paper, we propose an effective non-exemplar method called RAMF consisting of Random Auxiliary classes augmentation and Mixed Feature. On the one hand, we design a novel random auxiliary classes augmentation method, where one augmentation is randomly selected from three augmentations and applied on the input to generate augmented samples and extra class labels. By extending data and label space, it allows the model to learn more diverse representations, which can prevent the model from being biased towards learning task-specific features. When learning new tasks, it will reduce the change of feature space and improve model generalization. On the other hand, we employ mixed feature to replace the new features since only using new feature to optimize the model will affect the representation that was previously embedded in the feature space. Instead, by mixing new and old features, old knowledge can be retained without increasing the computational complexity. Extensive experiments on three benchmarks demonstrate the superiority of our approach, which outperforms the state-of-the-art non-exemplar methods and is comparable to high-performance replay-based methods.

arXiv:2304.08130

A Survey on Few-Shot Class-Incremental Learning

Large deep learning models are impressive, but they struggle when real-time data is not available. Few-shot class-incremental learning (FSCIL) poses a significant challenge for deep neural networks to learn new tasks from just a few labeled samples without forgetting the previously learned ones. This setup easily leads to catastrophic forgetting and overfitting problems, severely affecting model performance. Studying FSCIL helps overcome deep learning model limitations on data volume and acquisition time, while improving practicality and adaptability of machine learning models. This paper provides a comprehensive survey on FSCIL. Unlike previous surveys, we aim to synthesize few-shot learning and incremental learning, focusing on introducing FSCIL from two perspectives, while reviewing over 30 theoretical research studies and more than 20 applied research studies. From the theoretical perspective, we provide a novel categorization approach that divides the field into five subcategories, including traditional machine learning methods, meta-learning based methods, feature and feature space-based methods, replay-based methods, and dynamic network structure-based methods. We also evaluate the performance of recent theoretical research on benchmark datasets of FSCIL. From the application perspective, FSCIL has achieved impressive achievements in various fields of computer vision such as image classification, object detection, and image segmentation, as well as in natural language processing and graph. We summarize the important applications. Finally, we point out potential future research directions, including applications, problem setups, and theory development. Overall, this paper offers a comprehensive analysis of the latest advances in FSCIL from a methodological, performance, and application perspective.

arXiv:2304.06775

PointCLIMB: An Exemplar-Free Point Cloud Class Incremental Benchmark

Point clouds offer comprehensive and precise data regarding the contour and configuration of objects. Employing such geometric and topological 3D information of objects in class incremental learning can aid endless application in 3D-computer vision. Well known 3D-point cloud class incremental learning methods for addressing catastrophic forgetting generally entail the usage of previously encountered data, which can present difficulties in situations where there are restrictions on memory or when there are concerns about the legality of the data. Towards this we pioneer to leverage exemplar free class incremental learning on Point Clouds. In this paper we propose PointCLIMB: An exemplar Free Class Incremental Learning Benchmark. We focus on a pragmatic perspective to consider novel classes for class incremental learning on 3D point clouds. We setup a benchmark for 3D Exemplar free class incremental learning. We investigate performance of various backbones on 3D-Exemplar Free Class Incremental Learning framework. We demonstrate our results on ModelNet40 dataset.

arXiv:2304.06971공식 코드 · Papers with Code

Preserving Locality in Vision Transformers for Class Incremental Learning

Learning new classes without forgetting is crucial for real-world applications for a classification model. Vision Transformers (ViT) recently achieve remarkable performance in Class Incremental Learning (CIL). Previous works mainly focus on block design and model expansion for ViTs. However, in this paper, we find that when the ViT is incrementally trained, the attention layers gradually lose concentration on local features. We call this interesting phenomenon as \emph{Locality Degradation} in ViTs for CIL. Since the low-level local information is crucial to the transferability of the representation, it is beneficial to preserve the locality in attention layers. In this paper, we encourage the model to preserve more local information as the training procedure goes on and devise a Locality-Preserved Attention (LPA) layer to emphasize the importance of local features. Specifically, we incorporate the local information directly into the vanilla attention and control the initial gradients of the vanilla attention by weighting it with a small initial value. Extensive experiments show that the representations facilitated by LPA capture more low-level general information which is easier to transfer to follow-up tasks. The improved model gets consistently better performance on CIFAR100 and ImageNet100.

arXiv:2304.06619

Class-Incremental Learning of Plant and Disease Detection: Growing Branches with Knowledge Distillation

This paper investigates the problem of class-incremental object detection for agricultural applications where a model needs to learn new plant species and diseases incrementally without forgetting the previously learned ones. We adapt two public datasets to include new categories over time, simulating a more realistic and dynamic scenario. We then compare three class-incremental learning methods that leverage different forms of knowledge distillation to mitigate catastrophic forgetting. Our experiments show that all three methods suffer from catastrophic forgetting, but the Dynamic Y-KD approach, which additionally uses a dynamic architecture that grows new branches to learn new tasks, outperforms ILOD and Faster-ILOD in most settings both on new and old classes. These results highlight the challenges and opportunities of continual object detection for agricultural applications. In particular, we hypothesize that the large intra-class and small inter-class variability that is typical of plant images exacerbate the difficulty of learning new categories without interfering with previous knowledge. We publicly release our code to encourage future work.

arXiv:2304.05734

Few-shot Class-incremental Learning for Cross-domain Disease Classification

The ability to incrementally learn new classes from limited samples is crucial to the development of artificial intelligence systems for real clinical application. Although existing incremental learning techniques have attempted to address this issue, they still struggle with only few labeled data, particularly when the samples are from varied domains. In this paper, we explore the cross-domain few-shot incremental learning (CDFSCIL) problem. CDFSCIL requires models to learn new classes from very few labeled samples incrementally, and the new classes may be vastly different from the target space. To counteract this difficulty, we propose a cross-domain enhancement constraint and cross-domain data augmentation method. Experiments on MedMNIST show that the classification performance of this method is better than other similar incremental learning methods.

arXiv:2304.05547

Taxonomic Class Incremental Learning

The problem of continual learning has attracted rising attention in recent years. However, few works have questioned the commonly used learning setup, based on a task curriculum of random class. This differs significantly from human continual learning, which is guided by taxonomic curricula. In this work, we propose the Taxonomic Class Incremental Learning (TCIL) problem. In TCIL, the task sequence is organized based on a taxonomic class tree. We unify existing approaches to CIL and taxonomic learning as parameter inheritance schemes and introduce a new such scheme for the TCIL learning. This enables the incremental transfer of knowledge from ancestor to descendant class of a class taxonomy through parameter inheritance. Experiments on CIFAR-100 and ImageNet-100 show the effectiveness of the proposed TCIL method, which outperforms existing SOTA methods by 2% in terms of final accuracy on CIFAR-100 and 3% on ImageNet-100.

arXiv:2304.04854

iPINNs: Incremental learning for Physics-informed neural networks

Physics-informed neural networks (PINNs) have recently become a powerful tool for solving partial differential equations (PDEs). However, finding a set of neural network parameters that lead to fulfilling a PDE can be challenging and non-unique due to the complexity of the loss landscape that needs to be traversed. Although a variety of multi-task learning and transfer learning approaches have been proposed to overcome these issues, there is no incremental training procedure for PINNs that can effectively mitigate such training challenges. We propose incremental PINNs (iPINNs) that can learn multiple tasks (equations) sequentially without additional parameters for new tasks and improve performance for every equation in the sequence. Our approach learns multiple PDEs starting from the simplest one by creating its own subnetwork for each PDE and allowing each subnetwork to overlap with previously learned subnetworks. We demonstrate that previous subnetworks are a good initialization for a new equation if PDEs share similarities. We also show that iPINNs achieve lower prediction error than regular PINNs for two different scenarios: (1) learning a family of equations (e.g., 1-D convection PDE); and (2) learning PDEs resulting from a combination of processes (e.g., 1-D reaction-diffusion PDE). The ability to learn all problems with a single network together with learning more complex PDEs with better generalization than regular PINNs will open new avenues in this field.

arXiv:2304.05362

MASIL: Towards Maximum Separable Class Representation for Few Shot Class Incremental Learning

Few Shot Class Incremental Learning (FSCIL) with few examples per class for each incremental session is the realistic setting of continual learning since obtaining large number of annotated samples is not feasible and cost effective. We present the framework MASIL as a step towards learning the maximal separable classifier. It addresses the common problem i.e forgetting of old classes and over-fitting to novel classes by learning the classifier weights to be maximally separable between classes forming a simplex Equiangular Tight Frame. We propose the idea of concept factorization explaining the collapsed features for base session classes in terms of concept basis and use these to induce classifier simplex for few shot classes. We further adds fine tuning to reduce any error occurred during factorization and train the classifier jointly on base and novel classes without retaining any base class samples in memory. Experimental results on miniImageNet, CIFAR-100 and CUB-200 demonstrate that MASIL outperforms all the benchmarks.

arXiv:2304.05255

Density Map Distillation for Incremental Object Counting

We investigate the problem of incremental learning for object counting, where a method must learn to count a variety of object classes from a sequence of datasets. A na\"ive approach to incremental object counting would suffer from catastrophic forgetting, where it would suffer from a dramatic performance drop on previous tasks. In this paper, we propose a new exemplar-free functional regularization method, called Density Map Distillation (DMD). During training, we introduce a new counter head for each task and introduce a distillation loss to prevent forgetting of previous tasks. Additionally, we introduce a cross-task adaptor that projects the features of the current backbone to the previous backbone. This projector allows for the learning of new features while the backbone retains the relevant features for previous tasks. Finally, we set up experiments of incremental learning for counting new objects. Results confirm that our method greatly reduces catastrophic forgetting and outperforms existing methods.

arXiv:2304.04646

ECG-CL: A Comprehensive Electrocardiogram Interpretation Method Based on Continual Learning

Electrocardiogram (ECG) monitoring is one of the most powerful technique of cardiovascular disease (CVD) early identification, and the introduction of intelligent wearable ECG devices has enabled daily monitoring. However, due to the need for professional expertise in the ECGs interpretation, general public access has once again been restricted, prompting the need for the development of advanced diagnostic algorithms. Classic rule-based algorithms are now completely outperformed by deep learning based methods. But the advancement of smart diagnostic algorithms is hampered by issues like small dataset, inconsistent data labeling, inefficient use of local and global ECG information, memory and inference time consuming deployment of multiple models, and lack of information transfer between tasks. We propose a multi-resolution model that can sustain high-resolution low-level semantic information throughout, with the help of the development of low-resolution high-level semantic information, by capitalizing on both local morphological information and global rhythm information. From the perspective of effective data leverage and inter-task knowledge transfer, we develop a parameter isolation based ECG continual learning (ECG-CL) approach. We evaluated our model's performance on four open-access datasets by designing segmentation-to-classification for cross-domain incremental learning, minority-to-majority class for category incremental learning, and small-to-large sample for task incremental learning. Our approach is shown to successfully extract informative morphological and rhythmic features from ECG segmentation, leading to higher quality classification results. From the perspective of intelligent wearable applications, the possibility of a comprehensive ECG interpretation algorithm based on single-lead ECGs is also confirmed.

arXiv:2304.03980공식 코드 · Papers with Code

Continual Learning for LiDAR Semantic Segmentation: Class-Incremental and Coarse-to-Fine strategies on Sparse Data

During the last few years, continual learning (CL) strategies for image classification and segmentation have been widely investigated designing innovative solutions to tackle catastrophic forgetting, like knowledge distillation and self-inpainting. However, the application of continual learning paradigms to point clouds is still unexplored and investigation is required, especially using architectures that capture the sparsity and uneven distribution of LiDAR data. The current paper analyzes the problem of class incremental learning applied to point cloud semantic segmentation, comparing approaches and state-of-the-art architectures. To the best of our knowledge, this is the first example of class-incremental continual learning for LiDAR point cloud semantic segmentation. Different CL strategies were adapted to LiDAR point clouds and tested, tackling both classic fine-tuning scenarios and the Coarse-to-Fine learning paradigm. The framework has been evaluated through two different architectures on SemanticKITTI, obtaining results in line with state-of-the-art CL strategies and standard offline learning.

arXiv:2304.04222공식 코드 · Papers with Code

CILIATE: Towards Fairer Class-based Incremental Learning by Dataset and Training Refinement

Due to the model aging problem, Deep Neural Networks (DNNs) need updates to adjust them to new data distributions. The common practice leverages incremental learning (IL), e.g., Class-based Incremental Learning (CIL) that updates output labels, to update the model with new data and a limited number of old data. This avoids heavyweight training (from scratch) using conventional methods and saves storage space by reducing the number of old data to store. But it also leads to poor performance in fairness. In this paper, we show that CIL suffers both dataset and algorithm bias problems, and existing solutions can only partially solve the problem. We propose a novel framework, CILIATE, that fixes both dataset and algorithm bias in CIL. It features a novel differential analysis guided dataset and training refinement process that identifies unique and important samples overlooked by existing CIL and enforces the model to learn from them. Through this process, CILIATE improves the fairness of CIL by 17.03%, 22.46%, and 31.79% compared to state-of-the-art methods, iCaRL, BiC, and WA, respectively, based on our evaluation on three popular datasets and widely used ResNet models.

arXiv:2304.01899

Cross-Class Feature Augmentation for Class Incremental Learning

We propose a novel class incremental learning approach by incorporating a feature augmentation technique motivated by adversarial attacks. We employ a classifier learned in the past to complement training examples rather than simply play a role as a teacher for knowledge distillation towards subsequent models. The proposed approach has a unique perspective to utilize the previous knowledge in class incremental learning since it augments features of arbitrary target classes using examples in other classes via adversarial attacks on a previously learned classifier. By allowing the cross-class feature augmentations, each class in the old tasks conveniently populates samples in the feature space, which alleviates the collapse of the decision boundaries caused by sample deficiency for the previous tasks, especially when the number of stored exemplars is small. This idea can be easily incorporated into existing class incremental learning algorithms without any architecture modification. Extensive experiments on the standard benchmarks show that our method consistently outperforms existing class incremental learning methods by significant margins in various scenarios, especially under an environment with an extremely limited memory budget.

arXiv:2304.01663

On the Stability-Plasticity Dilemma of Class-Incremental Learning

A primary goal of class-incremental learning is to strike a balance between stability and plasticity, where models should be both stable enough to retain knowledge learned from previously seen classes, and plastic enough to learn concepts from new classes. While previous works demonstrate strong performance on class-incremental benchmarks, it is not clear whether their success comes from the models being stable, plastic, or a mixture of both. This paper aims to shed light on how effectively recent class-incremental learning algorithms address the stability-plasticity trade-off. We establish analytical tools that measure the stability and plasticity of feature representations, and employ such tools to investigate models trained with various algorithms on large-scale class-incremental benchmarks. Surprisingly, we find that the majority of class-incremental learning algorithms heavily favor stability over plasticity, to the extent that the feature extractor of a model trained on the initial set of classes is no less effective than that of the final incremental model. Our observations not only inspire two simple algorithms that highlight the importance of feature representation analysis, but also suggest that class-incremental learning approaches, in general, should strive for better feature representation learning.

arXiv:2304.00488

Saddle-to-Saddle Dynamics in Diagonal Linear Networks

In this paper we fully describe the trajectory of gradient flow over diagonal linear networks in the limit of vanishing initialisation. We show that the limiting flow successively jumps from a saddle of the training loss to another until reaching the minimum $\ell_1$-norm solution. This saddle-to-saddle dynamics translates to an incremental learning process as each saddle corresponds to the minimiser of the loss constrained to an active set outside of which the coordinates must be zero. We explicitly characterise the visited saddles as well as the jumping times through a recursive algorithm reminiscent of the LARS algorithm used for computing the Lasso path. Our proof leverages a convenient arc-length time-reparametrisation which enables to keep track of the heteroclinic transitions between the jumps. Our analysis requires negligible assumptions on the data, applies to both under and overparametrised settings and covers complex cases where there is no monotonicity of the number of active coordinates. We provide numerical experiments to support our findings.

arXiv:2304.00219

ConvBLS: An Effective and Efficient Incremental Convolutional Broad Learning System for Image Classification

Deep learning generally suffers from enormous computational resources and time-consuming training processes. Broad Learning System (BLS) and its convolutional variants have been proposed to mitigate these issues and have achieved superb performance in image classification. However, the existing convolutional-based broad learning system (C-BLS) either lacks an efficient training method and incremental learning capability or suffers from poor performance. To this end, we propose a convolutional broad learning system (ConvBLS) based on the spherical K-means (SKM) algorithm and two-stage multi-scale (TSMS) feature fusion, which consists of the convolutional feature (CF) layer, convolutional enhancement (CE) layer, TSMS feature fusion layer, and output layer. First, unlike the current C-BLS, the simple yet efficient SKM algorithm is utilized to learn the weights of CF layers. Compared with random filters, the SKM algorithm makes the CF layer learn more comprehensive spatial features. Second, similar to the vanilla BLS, CE layers are established to expand the feature space. Third, the TSMS feature fusion layer is proposed to extract more effective multi-scale features through the integration of CF layers and CE layers. Thanks to the above design and the pseudo-inverse calculation of the output layer weights, our proposed ConvBLS method is unprecedentedly efficient and effective. Finally, the corresponding incremental learning algorithms are presented for rapid remodeling if the model deems to expand. Experiments and comparisons demonstrate the superiority of our method.