To avoid collapse in self-supervised learning (SSL), a contrastive loss is widely used but often requires a large number of negative samples. Without negative samples yet achieving competitive performance, a recent work has attracted significant attention for providing a minimalist simple Siamese (SimSiam) method to avoid collapse. However, the reason for how it avoids collapse without negative samples remains not fully clear and our investigation starts by revisiting the explanatory claims in the original SimSiam. After refuting their claims, we introduce vector decomposition for analyzing the collapse based on the gradient analysis of the $l_2$-normalized representation vector. This yields a unified perspective on how negative samples and SimSiam alleviate collapse. Such a unified perspective comes timely for understanding the recent progress in SSL.
Carnegie Mellon University（卡内基梅隆大学）；University of California, Berkeley（加州大学伯克利分校）
Contrastive methods have led a recent surge in the performance of self-supervised representation learning (SSL). Recent methods like BYOL or SimSiam purportedly distill these contrastive methods down to their essence, removing bells and whistles, including the negative examples, that do not contribute to downstream performance. These "non-contrastive" methods work surprisingly well without using negatives even though the global minimum lies at trivial collapse. We empirically analyze these non-contrastive methods and find that SimSiam is extraordinarily sensitive to dataset and model size. In …
Ping An Technology (Shenzhen) Co., Ltd.（平安科技（深圳）有限公司）；School of Economics and Finance, Shanghai International Studies University（上海外国语大学经济与金融学院）
Nonparallel multi-domain voice conversion methods such as the StarGAN-VCs have been widely applied in many scenarios. However, the training of these models usually poses a challenge due to their complicated adversarial network architectures. To address this, in this work we leverage the state-of-the-art contrastive learning techniques and incorporate an efficient Siamese network structure into the StarGAN discriminator. Our method is called SimSiam-StarGAN-VC and it boosts the training stability and effectively prevents the discriminator overfitting issue in the training process. We conduct ex…
Behaviour biometrics are being explored as a viable alternative to overcome the limitations of traditional authentication methods such as passwords and static biometrics. Also, they are being considered as a viable authentication method for IoT devices such as smart headsets with AR/VR capabilities, wearables, and erables, that do not have a large form factor or the ability to seamlessly interact with the user. Recent behavioural biometric solutions use deep learning models that require large amounts of annotated training data. Collecting such volumes of behaviour biometrics data raises privac…
Facebook AI Research（脸书人工智能研究院）；Stanford University（斯坦福大学）
While contrastive approaches of self-supervised learning (SSL) learn representations by minimizing the distance between two augmented views of the same data point (positive pairs) and maximizing views from different data points (negative pairs), recent \emph{non-contrastive} SSL (e.g., BYOL and SimSiam) show remarkable performance {\it without} negative pairs, with an extra learnable predictor and a stop-gradient operation. A fundamental question arises: why do these methods not collapse into trivial representations? We answer this question via a simple theoretical study and propose a novel ap…
Duke University（杜克大学）；Facebook AI Research（脸书人工智能研究院）；University of Washington（华盛顿大学）
Non-contrastive methods of self-supervised learning (such as BYOL and SimSiam) learn representations by minimizing the distance between two views of the same image. These approaches have achieved remarkable performance in practice, but the theoretical understanding lags behind. Tian et al. 2021 explained why the representation does not collapse to zero, however, how the feature is learned still remains mysterious. In our work, we prove in a linear network, non-contrastive methods learn a desirable projection matrix and also reduce the sample complexity on downstream tasks. Our analysis suggest…
Residual Channel Boosts Contrastive Learning for Radio Frequency Fingerprint Identification
School of Information and Communication Engineering, University of Electronic Science and Technology of China（电子科技大学信息与通信工程学院）；School of Cyber Science and Engineering, Southeast University（东南大学网络空间安全学院）；Zhejiang Lab（之江实验室）
Recently, a variety of methods under the name of non-contrastive learning (like BYOL, SimSiam, SwAV, DINO) show that when equipped with some asymmetric architectural designs, aligning positive pairs alone is sufficient to attain good performance in self-supervised visual learning. Despite some understandings of some specific modules (like the predictor in BYOL), there is yet no unified theoretical understanding of how these seemingly different asymmetric designs can all avoid feature collapse, particularly considering methods that also work without the predictor (like DINO). In this work, we p…
Facebook AI Research（脸书人工智能研究院）；UNC Chapel Hill（北卡罗来纳大学教堂山分校）；Dartmouth（达特茅斯学院）
Video transformers have recently emerged as a competitive alternative to 3D CNNs for video understanding. However, due to their large number of parameters and reduced inductive biases, these models require supervised pretraining on large-scale image datasets to achieve top performance. In this paper, we empirically demonstrate that self-supervised pretraining of video transformers on video-only datasets can lead to action recognition results that are on par or better than those obtained with supervised pretraining on large-scale image datasets, even massive ones such as ImageNet-21K. Since tra…
Mila（米拉研究所）；Université de Montréal（蒙特利尔大学）；DeepMind（深度思维）；Microsoft Research（微软研究院）；Google Research（谷歌研究院）
Self-Supervised Learning (SSL) methods operate on unlabeled data to learn robust representations useful for downstream tasks. Most SSL methods rely on augmentations obtained by transforming the 2D image pixel map. These augmentations ignore the fact that biological vision takes place in an immersive three-dimensional, temporally contiguous environment, and that low-level biological vision relies heavily on depth cues. Using a signal provided by a pretrained state-of-the-art monocular RGB-to-depth model (the \emph{Depth Prediction Transformer}, Ranftl et al., 2021), we explore two distinct appr…
Xi’an Jiaotong-liverpool University（西交利物浦大学）；University of Science and Technology of China（中国科学技术大学）；Institute of Computing Technology, CAS（中国科学院计算技术研究所）
MIT CSAIL（麻省理工学院计算机科学与人工智能实验室）；MIT LIDS（麻省理工学院信息与决策系统实验室）；University of Pittsburgh（匹兹堡大学）
The generalization of representations learned via contrastive learning depends crucially on what features of the data are extracted. However, we observe that the contrastive loss does not always sufficiently guide which features are extracted, a behavior that can negatively impact the performance on downstream tasks via "shortcuts", i.e., by inadvertently suppressing important predictive features. We find that feature extraction is influenced by the difficulty of the so-called instance discrimination task (i.e., the task of discriminating pairs of similar points from pairs of dissimilar ones).…
In this report, we relate the algorithmic design of Barlow Twins' method to the Hilbert-Schmidt Independence Criterion (HSIC), thus establishing it as a contrastive learning approach that is free of negative samples. Through this perspective, we argue that Barlow Twins (and thus the class of negative-sample-free contrastive learning methods) suggests a possibility to bridge the two major families of self-supervised learning philosophies: non-contrastive and contrastive approaches. In particular, Barlow twins exemplified how we could combine the best practices of both worlds: avoiding the need …
Dual Perspectives on Non-Contrastive Self-Supervised Learning
Meta FAIR；New York University（纽约大学）；INRIA Paris（法国国家信息与自动化研究所巴黎中心）；Carnegie-Mellon University（卡内基梅隆大学）；Univ. Grenoble Alpes（格勒诺布尔阿尔卑斯大学）；Inria（法国国家信息与自动化研究所）；CNRS（法国国家科学研究中心）；Grenoble INP（格勒诺布尔国立高等物理、数学与信息学院）
Contrastive loss has significantly improved performance in supervised classification tasks by using a multi-viewed framework that leverages augmentation and label information. The augmentation enables contrast with another view of a single image but enlarges training time and memory usage. To exploit the strength of multi-views while avoiding the high computation cost, we introduce a multi-exit architecture that outputs multiple features of a single image in a single-viewed framework. To this end, we propose Self-Contrastive (SelfCon) learning, which self-contrasts within multiple outputs from…
DConAD: A Differencing-based Contrastive Representation Learning Framework for Time Series Anomaly Detection
University of Chinese Academy of Science（中国科学院大学）；Tsinghua University（清华大学）；Shanghai University of International Business and Economics（上海国际商务经济学院）；Northeast Normal University（东北师范大学）；Fuzhou University（福州市大学）；Hubei University（湖北大学）；Hainan University（海南大学）；Northeastern University（东北大学）
Albeit having gained significant progress lately, large-scale graph representation learning remains expensive to train and deploy for two main reasons: (i) the repetitive computation of multi-hop message passing and non-linearity in graph neural networks (GNNs); (ii) the computational cost of complex pairwise contrastive learning loss. Two main contributions are made in this paper targeting this twofold challenge: we first propose an adaptive-view graph neural encoder (AVGE) with a limited number of message passing to accelerate the forward pass computation, and then we propose a structure-awa…
A lot of recent advances in unsupervised feature learning are based on designing features which are invariant under semantic data augmentations. A common way to do this is contrastive learning, which uses positive and negative samples. Some recent works however have shown promising results for non-contrastive learning, which does not require negative samples. However, the non-contrastive losses have obvious "collapsed" minima, in which the encoders output a constant feature embedding, independent of the input. A folk conjecture is that so long as these collapsed solutions are avoided, the prod…
Self-supervised learning of graph neural networks (GNN) is in great need because of the widespread label scarcity issue in real-world graph/network data. Graph contrastive learning (GCL), by training GNNs to maximize the correspondence between the representations of the same graph in its different augmented forms, may yield robust and transferable GNNs even without using labels. However, GNNs trained by traditional GCL often risk capturing redundant graph features and thus may be brittle and provide sub-par performance in downstream tasks. Here, we propose a novel principle, termed adversarial…
Contrastive learning allows us to flexibly define powerful losses by contrasting positive pairs from sets of negative samples. Recently, the principle has also been used to learn cross-modal embeddings for video and text, yet without exploiting its full potential. In particular, previous losses do not take the intra-modality similarities into account, which leads to inefficient embeddings, as the same content is mapped to multiple points in the embedding space. With CrossCLR, we present a contrastive loss that fixes this issue. Moreover, we define sets of highly related samples in terms of the…
Contrastive learning has recently achieved remarkable success in many domains including graphs. However contrastive loss, especially for graphs, requires a large number of negative samples which is unscalable and computationally prohibitive with a quadratic time complexity. Sub-sampling is not optimal and incorrect negative sampling leads to sampling bias. In this work, we propose a meta-node based approximation technique that can (a) proxy all negative combinations (b) in quadratic cluster size time complexity, (c) at graph level, not node level, and (d) exploit graph sparsity. By replacing n…
Incomplete multi-view clustering (IMVC) is an unsupervised approach, among which IMVC via contrastive learning has received attention due to its excellent performance. The previous methods have the following problems: 1) Over-reliance on additional projection heads when solving the dimensional collapse problem in which latent features are only valid in lower-dimensional subspaces during clustering. However, many parameters in the projection heads are unnecessary. 2) The recovered view contain inconsistent private information and useless private information will mislead the learning of common s…
Preferred Networks, Inc.（Preferred Networks公司）；University of Oxford（牛津大学）
In contrastive representation learning, data representation is trained so that it can classify the image instances even when the images are altered by augmentations. However, depending on the datasets, some augmentations can damage the information of the images beyond recognition, and such augmentations can result in collapsed representations. We present a partial solution to this problem by formalizing a stochastic encoding process in which there exist a tug-of-war between the data corruption introduced by the augmentations and the information preserved by the encoder. We show that, with the …
Supervised Adversarial Contrastive Learning for Emotion Recognition in Conversations
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院）
Peking University（北京大学）；Tencent AI Lab（腾讯人工智能实验室）；Shandong University（山东大学）；Beijing Institute of General Artificial Intelligence（北京通用人工智能研究院）；Renmin University of China（中国人民大学）
Retrieval-augmented Neural Machine Translation models have been successful in many translation scenarios. Different from previous works that make use of mutually similar but redundant translation memories~(TMs), we propose a new retrieval-augmented NMT to model contrastively retrieved translation memories that are holistically similar to the source sentence while individually contrastive to each other providing maximal information gains in three phases. First, in TM retrieval phase, we adopt a contrastive retrieval algorithm to avoid redundancy and uninformativeness of similar translation piec…
Carnegie Mellon University（卡内基梅隆大学）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Contrastive learning has been applied successfully to learn vector representations of text. Previous research demonstrated that learning high-quality representations benefits from batch-wise contrastive loss with a large number of negatives. In practice, the technique of in-batch negative is used, where for each example in a batch, other batch examples' positives will be taken as its negatives, avoiding encoding extra negatives. This, however, still conditions each example's loss on all batch examples and requires fitting the entire large batch into GPU memory. This paper introduces a gradient…
Contrastive learning is among the most successful methods for visual representation learning, and its performance can be further improved by jointly performing clustering on the learned representations. However, existing methods for joint clustering and contrastive learning do not perform well on long-tailed data distributions, as majority classes overwhelm and distort the loss of minority classes, thus preventing meaningful representations to be learned. Motivated by this, we develop a novel joint clustering and contrastive learning framework by adapting the debiased contrastive loss to avoid…
CDT, University of Oxford（牛津大学CDT）；IDEA Research, International Digital Economy Academy（IDEA研究院、国际数字经济学院）；School of Advanced Interdisciplinary Sciences, UCAS（北京大学交叉学科研究院）；State Key Lab of AI Safety, Institute of Computing Technology, CAS（人工智能安全国家重点实验室，计算技术研究所，中国科学院）；China Mobile Research Institute（中国移动研究院）；DataArc Tech Ltd.（DataArc科技有限公司）
Amazon Web Services（亚马逊云科技）；UMass Amherst（马萨诸塞大学阿默斯特分校）；UCLA（加州大学洛杉矶分校）
Few-shot learning aims to transfer information from one task to enable generalization on novel tasks given a few examples. This information is present both in the domain and the class labels. In this work we investigate the complementary roles of these two sources of information by combining instance-discriminative contrastive learning and supervised learning in a single framework called Supervised Momentum Contrastive learning (SUPMOCO). Our approach avoids a problem observed in supervised learning where information in images not relevant to the task is discarded, which hampers their generali…
