Building rich machine learning datasets in a scalable manner often necessitates a crowd-sourced data collection pipeline. In this work, we use human studies to investigate the consequences of employing such a pipeline, focusing on the popular ImageNet dataset. We study how specific design choices in the ImageNet creation process impact the fidelity of the resulting dataset---including the introduction of biases that state-of-the-art models exploit. Our analysis pinpoints how a noisy data collection pipeline can lead to a systematic misalignment between the resulting benchmark and the real-world task it serves as a proxy for. Finally, our findings emphasize the need to augment our current model training and evaluation toolkit to take such misalignments into account. To facilitate further research, we release our refined ImageNet annotations at https://github.com/MadryLab/ImageNetMultiLabel.
University of Washington（华盛顿大学）；Google Research（谷歌研究院）；Allen Institute for AI（艾伦人工智能研究所）
Does progress on ImageNet transfer to real-world datasets? We investigate this question by evaluating ImageNet pre-trained models with varying accuracy (57% - 83%) on six practical image classification datasets. In particular, we study datasets collected with the goal of solving real-world tasks (e.g., classifying images from camera traps or satellites), as opposed to web-scraped benchmarks collected for comparing models. On multiple datasets, models with higher ImageNet accuracy do not consistently yield performance improvements. For certain tasks, interventions such as data augmentation impr…
We introduce several new datasets namely ImageNet-A/O and ImageNet-R as well as a synthetic environment and testing suite we called CAOS. ImageNet-A/O allow researchers to focus in on the blind spots remaining in ImageNet. ImageNet-R was specifically created with the intention of tracking robust representation as the representations are no longer simply natural but include artistic, and other renditions. The CAOS suite is built off of CARLA simulator which allows for the inclusion of anomalous objects and can create reproducible synthetic environment and scenes for testing robustness. All of t…
MIT CSAIL（麻省理工学院计算机科学与人工智能实验室）；Amazon Prime Video（亚马逊Prime Video）
The learning objective of vision-language approach of CLIP does not effectively account for the noisy many-to-many correspondences found in web-harvested image captioning datasets, which contributes to its compute and data inefficiency. To address this challenge, we introduce a novel training framework based on cross-modal contrastive learning that uses progressive self-distillation and soft image-text alignments to more efficiently learn robust representations from noisy data. Our model distills its own knowledge to dynamically generate soft-alignment targets for a subset of images and captio…
Deep learning thrives with large neural networks and large datasets. However, larger networks and larger datasets result in longer training times that impede research and development progress. Distributed synchronous SGD offers a potential solution to this problem by dividing SGD minibatches over a pool of parallel workers. Yet to make this scheme efficient, the per-worker workload must be large, which implies nontrivial growth in the SGD minibatch size. In this paper, we empirically show that on the ImageNet dataset large minibatches cause optimization difficulties, but when these are address…
Recently, differentiable search methods have made major progress in reducing the computational costs of neural architecture search. However, these approaches often report lower accuracy in evaluating the searched architecture or transferring it to another dataset. This is arguably due to the large gap between the architecture depths in search and evaluation scenarios. In this paper, we present an efficient algorithm which allows the depth of searched architectures to grow gradually during the training procedure. This brings two issues, namely, heavier computational overheads and weaker search …
Representation learning promises to unlock deep learning for the long tail of vision tasks without expensive labelled datasets. Yet, the absence of a unified evaluation for general visual representations hinders progress. Popular protocols are often too constrained (linear classification), limited in diversity (ImageNet, CIFAR, Pascal-VOC), or only weakly related to representation quality (ELBO, reconstruction error). We present the Visual Task Adaptation Benchmark (VTAB), which defines good representations as those that adapt to diverse, unseen tasks with few examples. With VTAB, we conduct a…
LAION-C: An Out-of-Distribution Benchmark for Web-Scale Vision Models
Max Planck Institute for Intelligent Systems, Tübingen, Germany（马克斯·普朗克智能系统研究所，图宾根，德国）；ELLIS Institute Tübingen（图宾根ELLIS研究所）；Tübingen AI Center（图宾根人工智能中心）；Google DeepMind（谷歌DeepMind）
Cornell University（康奈尔大学）；Caltech（加州理工学院）；Google（谷歌）；University of Edinburgh（爱丁堡大学）
Recent progress in self-supervised learning has resulted in models that are capable of extracting rich representations from image collections without requiring any explicit label supervision. However, to date the vast majority of these approaches have restricted themselves to training on standard benchmark datasets such as ImageNet. We argue that fine-grained visual categorization problems, such as plant and animal species classification, provide an informative testbed for self-supervised learning. In order to facilitate progress in this area we present two new natural world visual classificat…
Invariance to a broad array of image corruptions, such as warping, noise, or color shifts, is an important aspect of building robust models in computer vision. Recently, several new data augmentations have been proposed that significantly improve performance on ImageNet-C, a benchmark of such corruptions. However, there is still a lack of basic understanding on the relationship between data augmentations and test-time corruptions. To this end, we develop a feature space for image transforms, and then use a new measure in this space between augmentations and corruptions called the Minimal Sampl…
Nearest neighbors is a successful and long-standing technique for anomaly detection. Significant progress has been recently achieved by self-supervised deep methods (e.g. RotNet). Self-supervised features however typically under-perform Imagenet pre-trained features. In this work, we investigate whether the recent progress can indeed outperform nearest-neighbor methods operating on an Imagenet pretrained feature space. The simple nearest-neighbor based-approach is experimentally shown to outperform self-supervised methods in: accuracy, few shot generalization, training time and noise robustnes…
Despite recent progress in computer vision, finegrained interpretation of satellite images remains challenging because of a lack of labeled training data. To overcome this limitation, we construct a novel dataset called WikiSatNet by pairing georeferenced Wikipedia articles with satellite imagery of their corresponding locations. We then propose two strategies to learn representations of satellite images by predicting properties of the corresponding articles from the images. Leveraging this new multi-modal dataset, we can drastically reduce the quantity of human-annotated labels and time requi…
Spanning Training Progress: Temporal Dual-Depth Scoring (TDDS) for Enhanced Dataset Pruning
XiDian University（西安电子科技大学）；Centre for Frontier AI Research (CFAR), Agency for Science, Technology and Research (A*STAR)（新加坡科技研究局前沿人工智能研究中心）；Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR)（新加坡科技研究局高性能计算研究所）
Efficient search is a core issue in Neural Architecture Search (NAS). It is difficult for conventional NAS algorithms to directly search the architectures on large-scale tasks like ImageNet. In general, the cost of GPU hours for NAS grows with regard to training dataset size and candidate set size. One common way is searching on a smaller proxy dataset (e.g., CIFAR-10) and then transferring to the target task (e.g., ImageNet). These architectures optimized on proxy data are not guaranteed to be optimal on the target task. Another common way is learning with a smaller candidate set, which may r…
Quantization has been applied to multiple domains in Deep Neural Networks (DNNs). We propose Depthwise Quantization (DQ) where $\textit{quantization}$ is applied to a decomposed sub-tensor along the $\textit{feature axis}$ of weak statistical dependence. The feature decomposition leads to an exponential increase in $\textit{representation capacity}$ with a linear increase in memory and parameter cost. In addition, DQ can be directly applied to existing encoder-decoder frameworks without modification of the DNN architecture. We use DQ in the context of Hierarchical Auto-Encoder and train end-to…
School of Geosciences and Info-Physics, Central South University（中南大学地球科学与信息物理学院）
In recent years, deep convolutional neural network (DCNN) has seen a breakthrough progress in natural image recognition because of three points: universal approximation ability via DCNN, large-scale database (such as ImageNet), and supercomputing ability powered by GPU. The remote sensing field is still lacking a large-scale benchmark compared to ImageNet and Place2. In this paper, we propose a remote sensing image classification benchmark (RSI-CB) based on massive, scalable, and diverse crowdsource data. Using crowdsource data, such as Open Street Map (OSM) data, ground objects in remote sens…
XNOR AI（XNOR人工智能公司）；University of Washington（华盛顿大学）；Allen AI（艾伦人工智能研究所）
Among the three main components (data, labels, and models) of any supervised learning system, data and models have been the main subjects of active research. However, studying labels and their properties has received very little attention. Current principles and paradigms of labeling impose several challenges to machine learning algorithms. Labels are often incomplete, ambiguous, and redundant. In this paper we study the effects of various properties of labels and introduce the Label Refinery: an iterative procedure that updates the ground truth labels after examining the entire dataset. We sh…
The existing few-shot video classification methods often employ a meta-learning paradigm by designing customized temporal alignment module for similarity calculation. While significant progress has been made, these methods fail to focus on learning effective representations, and heavily rely on the ImageNet pre-training, which might be unreasonable for the few-shot recognition setting due to semantics overlap. In this paper, we aim to present an in-depth study on few-shot video classification by making three contributions. First, we perform a consistent comparative study on the existing metric…
Google Research（谷歌研究院）；University of California Berkeley（加州大学伯克利分校）
Image classification accuracy on the ImageNet dataset has been a barometer for progress in computer vision over the last decade. Several recent papers have questioned the degree to which the benchmark remains useful to the community, yet innovations continue to contribute gains to performance, with today's largest models achieving 90%+ top-1 accuracy. To help contextualize progress on ImageNet and provide a more meaningful evaluation for today's state-of-the-art models, we manually review and categorize every remaining mistake that a few top models make in order to provide insight into the lon…
Yes, and no. We ask whether recent progress on the ImageNet classification benchmark continues to represent meaningful generalization, or whether the community has started to overfit to the idiosyncrasies of its labeling procedure. We therefore develop a significantly more robust procedure for collecting human annotations of the ImageNet validation set. Using these new labels, we reassess the accuracy of recently proposed ImageNet classifiers, and find their gains to be substantially smaller than those reported on the original labels. Furthermore, we find the original ImageNet labels to no lon…
The Evolution of Dataset Distillation: Toward Scalable and Generalizable Solutions
Department of Computer Science and Engineering, University of Nevada, Reno（计算机科学与工程系，内华达大学雷诺分校）；Centre for Frontier AI Research (CFAR), and Institute of High Performance Computing (IHPC), A*STAR, Singapore（前沿人工智能研究中心（CFAR）和高性能计算研究所（IHPC），A*STAR，新加坡）
We investigate the robustness properties of ResNeXt class image recognition models trained with billion scale weakly supervised data (ResNeXt WSL models). These models, recently made public by Facebook AI, were trained with ~1B images from Instagram and fine-tuned on ImageNet. We show that these models display an unprecedented degree of robustness against common image corruptions and perturbations, as measured by the ImageNet-C and ImageNet-P benchmarks. They also achieve substantially improved accuracies on the recently introduced "natural adversarial examples" benchmark (ImageNet-A). The lar…
Self-supervised learning holds promise in leveraging large amounts of unlabeled data, however much of its progress has thus far been limited to highly curated pre-training data such as ImageNet. We explore the effects of contrastive learning from larger, less-curated image datasets such as YFCC, and find there is indeed a large difference in the resulting representation quality. We hypothesize that this curation gap is due to a shift in the distribution of image classes -- which is more diverse and heavy-tailed -- resulting in less relevant negative samples to learn from. We test this hypothes…
University of Cambridge（剑桥大学）；University of Copenhagen（哥本哈根大学）；Mila – Quebec Artificial Intelligence Institute（米拉-魁北克人工智能研究所）；McGill University（麦吉尔大学）
The design of widespread vision-and-language datasets and pre-trained encoders directly adopts, or draws inspiration from, the concepts and images of ImageNet. While one can hardly overestimate how much this benchmark contributed to progress in computer vision, it is mostly derived from lexical databases and image queries in English, resulting in source material with a North American or Western European bias. Therefore, we devise a new protocol to construct an ImageNet-style hierarchy representative of more languages and cultures. In particular, we let the selection of both concepts and images…
Due to the scarcity of labeled data, using supervised models pre-trained on ImageNet is a de facto standard in remote sensing scene classification. Recently, the availability of larger high resolution remote sensing (HRRS) image datasets and progress in self-supervised learning have brought up the questions of whether supervised ImageNet pre-training is still necessary for remote sensing scene classification and would supervised pre-training on HRRS image datasets or self-supervised pre-training on ImageNet achieve better results on target remote sensing scene classification tasks. To answer t…
Hasso Plattner Institute（哈索·普拉特纳研究所）；University of Potsdam（波茨坦大学）
Binary Neural Networks (BNNs) show promising progress in reducing computational and memory costs but suffer from substantial accuracy degradation compared to their real-valued counterparts on large-scale datasets, e.g., ImageNet. Previous work mainly focused on reducing quantization errors of weights and activations, whereby a series of approximation methods and sophisticated training tricks have been proposed. In this work, we make several observations that challenge conventional wisdom. We revisit some commonly used techniques, such as scaling factors and custom gradients, and show that thes…
Three factors drive the advance of AI: algorithmic innovation, data, and the amount of compute available for training. Algorithmic progress has traditionally been more difficult to quantify than compute and data. In this work, we argue that algorithmic progress has an aspect that is both straightforward to measure and interesting: reductions over time in the compute needed to reach past capabilities. We show that the number of floating-point operations required to train a classifier to AlexNet-level performance on ImageNet has decreased by a factor of 44x between 2012 and 2019. This correspond…
National Institute of Advanced Industrial Science and Technology (AIST)（日本产业技术综合研究所（AIST））
The purpose of this study is to determine whether current video datasets have sufficient data for training very deep convolutional neural networks (CNNs) with spatio-temporal three-dimensional (3D) kernels. Recently, the performance levels of 3D CNNs in the field of action recognition have improved significantly. However, to date, conventional research has only explored relatively shallow 3D architectures. We examine the architectures of various 3D CNNs from relatively shallow to very deep ones on current video datasets. Based on the results of those experiments, the following conclusions coul…
We study the impact of using rich and diverse textual descriptions of classes for zero-shot learning (ZSL) on ImageNet. We create a new dataset ImageNet-Wiki that matches each ImageNet class to its corresponding Wikipedia article. We show that merely employing these Wikipedia articles as class descriptions yields much higher ZSL performance than prior works. Even a simple model using this type of auxiliary data outperforms state-of-the-art models that rely on standard features of word embedding encodings of class names. These results highlight the usefulness and importance of textual descripti…
University of Missouri-Columbia（密苏里大学哥伦比亚分校）；Beijing Jiaotong University（北京交通大学）
The inference structures and computational complexity of existing deep neural networks, once trained, are fixed and remain the same for all test images. However, in practice, it is highly desirable to establish a progressive structure for deep neural networks which is able to adapt its inference process and complexity for images with different visual recognition complexity. In this work, we develop a multi-stage progressive structure with integrated confidence analysis and decision policy learning for deep neural networks. This new framework consists of a set of network units to be activated i…
Yitu Tech（依图科技）；National University of Singapore（新加坡国立大学）；YITU Technology Institute of Data Science（依图科技数据科学研究所）
Transformers, which are popular for language modeling, have been explored for solving vision tasks recently, e.g., the Vision Transformer (ViT) for image classification. The ViT model splits each image into a sequence of tokens with fixed length and then applies multiple Transformer layers to model their global relation for classification. However, ViT achieves inferior performance to CNNs when trained from scratch on a midsize dataset like ImageNet. We find it is because: 1) the simple tokenization of input images fails to model the important local structure such as edges and lines among neig…
UC Berkeley（加州大学伯克利分校）；Bosch Research North America（博世北美研究院）
Recent progress in network-based audio event classification has shown the benefit of pre-training models on visual data such as ImageNet. While this process allows knowledge transfer across different domains, training a model on large-scale visual datasets is time consuming. On several audio event classification benchmarks, we show a fast and effective alternative that pre-trains the model unsupervised, only on audio data and yet delivers on-par performance with ImageNet pre-training. Furthermore, we show that our discriminative audio learning can be used to transfer knowledge across audio dat…
University of California San Diego（加利福尼亚大学圣迭戈分校）；CUNY Graduate Center（纽约市立大学研究生中心）；Facebook AI Research（Facebook人工智能研究院）
Recent progress in self-supervised learning has demonstrated promising results in multiple visual tasks. An important ingredient in high-performing self-supervised methods is the use of data augmentation by training models to place different augmented views of the same image nearby in embedding space. However, commonly used augmentation pipelines treat images holistically, ignoring the semantic relevance of parts of an image-e.g. a subject vs. a background-which can lead to the learning of spurious correlations. Our work addresses this problem by investigating a class of simple, yet highly eff…
National University of Defense Technology（国防科技大学）；Chinese Academy of Sciences（中国科学院）
Practical networks for edge devices adopt shallow depth and small convolutional kernels to save memory and computational cost, which leads to a restricted receptive field. Conventional efficient learning methods focus on lightweight convolution designs, ignoring the role of the receptive field in neural network design. In this paper, we propose the Meta-Pooling framework to make the receptive field learnable for a lightweight network, which consists of parameterized pooling-based operations. Specifically, we introduce a parameterized spatial enhancer, which is composed of pooling operations to…
Test sets are an integral part of evaluating models and gauging progress in object recognition, and more broadly in computer vision and AI. Existing test sets for object recognition, however, suffer from shortcomings such as bias towards the ImageNet characteristics and idiosyncrasies (e.g., ImageNet-V2), being limited to certain types of stimuli (e.g., indoor scenes in ObjectNet), and underestimating the model performance (e.g., ImageNet-A). To mitigate these problems, we introduce a new test set, called D2O, which is sufficiently different from existing test sets. Images are a mix of generat…
University of Tübingen（图宾根大学）；Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）
Computer graphics has experienced a recent surge of data-centric approaches for photorealistic and controllable content creation. StyleGAN in particular sets new standards for generative modeling regarding image quality and controllability. However, StyleGAN's performance severely degrades on large unstructured datasets such as ImageNet. StyleGAN was designed for controllability; hence, prior works suspect its restrictive design to be unsuitable for diverse datasets. In contrast, we find the main limiting factor to be the current training strategy. Following the recently introduced Projected G…
Empowered by large datasets, e.g., ImageNet, unsupervised learning on large-scale data has enabled significant advances for classification tasks. However, whether the large-scale unsupervised semantic segmentation can be achieved remains unknown. There are two major challenges: i) we need a large-scale benchmark for assessing algorithms; ii) we need to develop methods to simultaneously learn category and shape representation in an unsupervised manner. In this work, we propose a new problem of large-scale unsupervised semantic segmentation (LUSS) with a newly created benchmark dataset to help t…
Coarse-to-Fine: Progressive Image Compression for Semantically Hierarchical Classification
School of Integrated Technology, Yonsei University（延世大学整合技术学院）；BK21 Graduate Program in Intelligent Semiconductor Technology, Yonsei University（延世大学智能半导体技术研究生项目）；Department of Artificial Intelligence, Chung-Ang University（成均馆大学人工智能系）
Advancement in digital pathology and artificial intelligence has enabled deep learning-based computer vision techniques for automated disease diagnosis and prognosis. However, WSIs present unique computational and algorithmic challenges. WSIs are gigapixel-sized, making them infeasible to be used directly for training deep neural networks. Hence, for modeling, a two-stage approach is adopted: Patch representations are extracted first, followed by the aggregation for WSI prediction. These approaches require detailed pixel-level annotations for training the patch encoder. However, obtaining thes…
Unsupervised learning has made substantial progress over the last few years, especially by means of contrastive self-supervised learning. The dominating dataset for benchmarking self-supervised learning has been ImageNet, for which recent methods are approaching the performance achieved by fully supervised training. The ImageNet dataset is however largely object-centric, and it is not clear yet what potential those methods have on widely different datasets and tasks that are not object-centric, such as in digital pathology. While self-supervised learning has started to be explored within this …
Friedrich Schiller University Jena（耶拿弗里德里希·席勒大学）；Sapienza University of Rome（罗马第一大学）
Learning from limited amounts of data is the hallmark of intelligence, requiring strong generalization and abstraction skills. In a machine learning context, data-efficient methods are of high practical importance since data collection and annotation are prohibitively expensive in many domains. Thus, coordinated efforts to foster progress in this area emerged recently, e.g., in the form of dedicated workshops and competitions. Besides a common benchmark, measuring progress requires strong baselines. We present such a strong baseline for data-efficient image classification on the VIPriors chall…
Big neural networks trained on large datasets have advanced the state-of-the-art for a large variety of challenging problems, improving performance by a large margin. However, under low memory and limited computational power constraints, the accuracy on the same problems drops considerable. In this paper, we propose a series of techniques that significantly improve the accuracy of binarized neural networks (i.e networks where both the features and the weights are binary). We evaluate the proposed improvements on two diverse tasks: fine-grained recognition (human pose estimation) and large-scal…
FUJIFILM Corporation（富士胶片公司）；AWS AI Labs（AWS人工智能实验室）；Massachusetts Institute of Technology（麻省理工学院）；Stanford University（斯坦福大学）
Learning visual representations of medical images (e.g., X-rays) is core to medical image understanding but its progress has been held back by the scarcity of human annotations. Existing work commonly relies on fine-tuning weights transferred from ImageNet pretraining, which is suboptimal due to drastically different image characteristics, or rule-based label extraction from the textual report data paired with medical images, which is inaccurate and hard to generalize. Meanwhile, several recent studies show exciting results from unsupervised contrastive learning from natural images, but we fin…
Beyond Accuracy: Assessing Calibration of Geospatial Foundation Models and Their Sensitivity to Distribution Shifts
Technical University of Munich (TUM)（慕尼黑工业大学）；German Aerospace Center (DLR)（德国航空航天中心）；Massachusetts Institute of Technology (MIT)（麻省理工学院）；Taylor Geospatial（泰勒地理空间公司）
