We introduce Bootstrap Your Own Latent (BYOL), a new approach to self-supervised image representation learning. BYOL relies on two neural networks, referred to as online and target networks, that interact and learn from each other. From an augmented view of an image, we train the online network to predict the target network representation of the same image under a different augmented view. At the same time, we update the target network with a slow-moving average of the online network. While state-of-the art methods rely on negative pairs, BYOL achieves a new state of the art without them. BYOL reaches 74.3% top-1 classification accuracy on ImageNet using a linear evaluation with a ResNet-50 architecture and 79.6% with a larger ResNet. We show that BYOL performs on par or better than the current state of the art on both transfer and semi-supervised benchmarks. Our implementation and pretrained models are given on GitHub. 3 * Equal contribution; the order of first authors was randomly selected.

We introduce Bootstrap Your Own Latent (BYOL), a new approach to selfsupervised image representation learning. BYOL relies on two neural networks, referred to as online and target networks, that interact and learn from each other. From an augmented view of an image, we train the online network to predict the target network representation of the same image under a different augmented view. At the same time, we update the target network with a slow-moving average of the online network. While state-of-the art methods rely on negative pairs, BYOL achieves a new state of the art without them. BYOL reaches 74.3% top-1 classification accuracy on ImageNet using a linear evaluation with a ResNet-50 architecture and 79.6% with a larger ResNet. We show that BYOL performs on par or better than the current state of the art on both transfer and semi-supervised benchmarks. Our implementation and pretrained models are given on GitHub. 3 * Equal contribution; the order of first authors was randomly selected. 3

学习概括不见于没有人类监督的有效视觉表现是一个基本问题，以便将机器学习施加到各种各样的任务。最近，分别是SIMCLR和BYOL的两个自我监督方法，对比学习和潜在自动启动的家庭取得了重大进展。在这项工作中，我们假设向这些算法添加显式信息压缩产生更好，更强大的表示。我们通过开发与条件熵瓶颈（CEB）目标兼容的SIMCLR和BYOL配方来验证这一点，允许我们衡量并控制学习的表示中的压缩量，并观察它们对下游任务的影响。此外，我们探讨了Lipschitz连续性和压缩之间的关系，显示了我们学习的编码器的嘴唇峰常数上的易触摸下限。由于Lipschitz连续性与稳健性密切相关，这为什么压缩模型更加强大提供了新的解释。我们的实验证实，向SIMCLR和BYOL添加压缩显着提高了线性评估精度和模型鲁棒性，跨各种域移位。特别是，Byol的压缩版本与Reset-50的ImageNet上的76.0％的线性评估精度达到了76.0％的直线评价精度，并使用Reset-50 2x的78.8％。

This paper presents SimCLR: a simple framework for contrastive learning of visual representations. We simplify recently proposed contrastive selfsupervised learning algorithms without requiring specialized architectures or a memory bank. In order to understand what enables the contrastive prediction tasks to learn useful representations, we systematically study the major components of our framework. We show that (1) composition of data augmentations plays a critical role in defining effective predictive tasks, (2) introducing a learnable nonlinear transformation between the representation and the contrastive loss substantially improves the quality of the learned representations, and (3) contrastive learning benefits from larger batch sizes and more training steps compared to supervised learning. By combining these findings, we are able to considerably outperform previous methods for self-supervised and semi-supervised learning on ImageNet. A linear classifier trained on self-supervised representations learned by Sim-CLR achieves 76.5% top-1 accuracy, which is a 7% relative improvement over previous state-ofthe-art, matching the performance of a supervised ResNet-50. When fine-tuned on only 1% of the labels, we achieve 85.8% top-5 accuracy, outperforming AlexNet with 100× fewer labels. 1

尽管最近通过剩余网络的代表学习中的自我监督方法取得了进展，但它们仍然对ImageNet分类基准进行了高度的监督学习，限制了它们在性能关键设置中的适用性。在MITROVIC等人的现有理论上洞察中建立2021年，我们提出了RELICV2，其结合了明确的不变性损失，在各种适当构造的数据视图上具有对比的目标。 Relicv2在ImageNet上实现了77.1％的前1个分类准确性，使用线性评估使用Reset50架构和80.6％，具有较大的Reset型号，优于宽边缘以前的最先进的自我监督方法。最值得注意的是，RelicV2是使用一系列标准Reset架构始终如一地始终优先于类似的对比较中的监督基线的第一个表示学习方法。最后，我们表明，尽管使用Reset编码器，Relicv2可与最先进的自我监控视觉变压器相媲美。

许多最近的自我监督学习方法在图像分类和其他任务上表现出了令人印象深刻的表现。已经使用了一种令人困惑的多种技术，并不总是清楚地了解其收益的原因，尤其是在组合使用时。在这里，我们将图像的嵌入视为点粒子，并将模型优化视为该粒子系统上的动态过程。我们的动态模型结合了类似图像的吸引力，避免局部崩溃的局部分散力以及实现颗粒的全球均匀分布的全局分散力。动态透视图突出了使用延迟参数图像嵌入（a la byol）以及同一图像的多个视图的优点。它还使用纯动态的局部分散力（布朗运动），该分散力比其他方法显示出改善的性能，并且不需要其他粒子坐标的知识。该方法称为MSBREG，代表（i）多视质心损失，它施加了吸引力的力来将不同的图像视图嵌入到其质心上，（ii）奇异值损失，将粒子系统推向空间均匀的密度（ iii）布朗扩散损失。我们评估MSBREG在ImageNet上的下游分类性能以及转移学习任务，包括细粒度分类，多类对象分类，对象检测和实例分段。此外，我们还表明，将我们的正则化术语应用于其他方法，进一步改善了其性能并通过防止模式崩溃来稳定训练。

Self-supervised learning (SSL) is rapidly closing BARLOW TWINS is competitive with state-of-the-art methods for self-supervised learning while being conceptually simpler, naturally avoiding trivial constant (i.e. collapsed) embeddings, and being robust to the training batch size.

特征回归是将大型神经网络模型蒸馏到较小的功能回归。我们表明，随着网络架构的简单变化，回归可能会优于自我监督模型的知识蒸馏更复杂的最先进方法。令人惊讶的是，即使仅在蒸馏过程中仅使用并且在下游任务中丢弃时，将多层的Perceptron头部添加到CNN骨架上是有益的。因此，更深的非线性投影可以使用在不改变推理架构和时间的情况下准确地模仿老师。此外，我们利用独立的投影头来同时蒸馏多个教师网络。我们还发现，使用与教师和学生网络的输入相同的弱增强图像辅助蒸馏。Imagenet DataSet上的实验证明了各种自我监督蒸馏环境中提出的变化的功效。

Unsupervised image representations have significantly reduced the gap with supervised pretraining, notably with the recent achievements of contrastive learning methods. These contrastive methods typically work online and rely on a large number of explicit pairwise feature comparisons, which is computationally challenging. In this paper, we propose an online algorithm, SwAV, that takes advantage of contrastive methods without requiring to compute pairwise comparisons. Specifically, our method simultaneously clusters the data while enforcing consistency between cluster assignments produced for different augmentations (or "views") of the same image, instead of comparing features directly as in contrastive learning. Simply put, we use a "swapped" prediction mechanism where we predict the code of a view from the representation of another view. Our method can be trained with large and small batches and can scale to unlimited amounts of data. Compared to previous contrastive methods, our method is more memory efficient since it does not require a large memory bank or a special momentum network. In addition, we also propose a new data augmentation strategy, multi-crop, that uses a mix of views with different resolutions in place of two full-resolution views, without increasing the memory or compute requirements. We validate our findings by achieving 75.3% top-1 accuracy on ImageNet with ResNet-50, as well as surpassing supervised pretraining on all the considered transfer tasks.

我们提出了自适应培训 - 一种统一的培训算法，通过模型预测动态校准并增强训练过程，而不会产生额外的计算成本 - 以推进深度神经网络的监督和自我监督的学习。我们分析了培训数据的深网络培训动态，例如随机噪声和对抗例。我们的分析表明，模型预测能够在数据中放大有用的基础信息，即使在没有任何标签信息的情况下，这种现象也会发生，突出显示模型预测可能会产生培训过程：自适应培训改善了深网络的概括在噪音下，增强自我监督的代表学习。分析还阐明了解深度学习，例如，在经验风险最小化和最新的自我监督学习算法的折叠问题中对最近发现的双重现象的潜在解释。在CIFAR，STL和Imagenet数据集上的实验验证了我们在三种应用中的方法的有效性：用标签噪声，选择性分类和线性评估进行分类。为了促进未来的研究，该代码已在HTTPS://github.com/layneh/Self-Aveptive-训练中公开提供。

尽管增加了大量的增强家庭，但只有几个樱桃采摘的稳健增强政策有利于自我监督的图像代表学习。在本文中，我们提出了一个定向自我监督的学习范式（DSSL），其与显着的增强符号兼容。具体而言，我们在用标准增强的视图轻度增强后调整重增强策略，以产生更难的视图（HV）。 HV通常具有与原始图像较高的偏差而不是轻度增强的标准视图（SV）。与以前的方法不同，同等对称地将所有增强视图对称地最大化它们的相似性，DSSL将相同实例的增强视图视为部分有序集（具有SV $ \ LeftrightArrow $ SV，SV $ \左路$ HV），然后装备一个定向目标函数尊重视图之间的衍生关系。 DSSL可以轻松地用几行代码实现，并且对于流行的自我监督学习框架非常灵活，包括SIMCLR，Simsiam，Byol。对CiFar和Imagenet的广泛实验结果表明，DSSL可以稳定地改善各种基线，其兼容性与更广泛的增强。

我们从统计依赖性角度接近自我监督的图像表示学习，提出与希尔伯特 - 施密特独立性标准（SSL-HSIC）自我监督的学习。 SSL-HSIC最大化图像和图像标识的变换表示之间的依赖性，同时最小化这些表示的核化方差。该框架产生了对Infonce的新了解，在不同转换之间的相互信息（MI）上的变分下限。虽然已知MI本身具有可能导致学习无意义的表示的病理学，但其绑定表现得更好：我们表明它隐含地近似于SSL-HSIC（具有略微不同的规范器）。我们的方法还向我们深入了解Byol，一种无与伦比的SSL方法，因为SSL-HSIC类似地了解了当地的样本邻居。 SSL-HSIC允许我们在批量大小中直接在时间线性上直接优化统计依赖性，而无需限制数据假设或间接相互信息估计。 SSL-HSIC培训或没有目标网络，SSL-HSIC与Imagenet的标准线性评估相匹配，半监督学习和转移到其他分类和视觉任务，如语义分割，深度估计和对象识别等。代码可在https://github.com/deepmind/ssl_hsic提供。

我们专注于更好地理解增强不变代表性学习的关键因素。我们重新访问moco v2和byol，并试图证明以下假设的真实性：不同的框架即使具有相同的借口任务也会带来不同特征的表示。我们建立了MoCo V2和BYOL之间公平比较的第一个基准，并观察：（i）复杂的模型配置使得可以更好地适应预训练数据集； （ii）从实现竞争性转移表演中获得的预训练和微调阻碍模型的优化策略不匹配。鉴于公平的基准，我们进行进一步的研究并发现网络结构的不对称性赋予对比框架在线性评估协议下正常工作，同时可能会损害长尾分类任务的转移性能。此外，负样本并不能使模型更明智地选择数据增强，也不会使不对称网络结构结构。我们相信我们的发现为将来的工作提供了有用的信息。

近年来，基于对比的自我监督学习方法取得了巨大的成功。但是，自学要求非常长的训练时期（例如，MoCO V3的800个时代）才能获得有希望的结果，这对于一般学术界来说是不可接受的，并阻碍了该主题的发展。这项工作重新审视了基于动量的对比学习框架，并确定了两种增强观点仅产生一个积极对的效率低下。我们提出了快速MOCO-一个新颖的框架，该框架利用组合贴片从两个增强视图中构造了多对正面，该视图提供了丰富的监督信号，这些信号带来了可忽视的额外计算成本，从而带来了显着的加速。经过100个时期训练的快速MOCO实现了73.5％的线性评估精度，类似于经过800个时期训练的MOCO V3（Resnet-50骨干）。额外的训练（200个时期）进一步将结果提高到75.1％，这与最先进的方法相当。几个下游任务的实验也证实了快速MOCO的有效性。

Contrastive representation learning has proven to be an effective self-supervised learning method for images and videos. Most successful approaches are based on Noise Contrastive Estimation (NCE) and use different views of an instance as positives that should be contrasted with other instances, called negatives, that are considered as noise. However, several instances in a dataset are drawn from the same distribution and share underlying semantic information. A good data representation should contain relations between the instances, or semantic similarity and dissimilarity, that contrastive learning harms by considering all negatives as noise. To circumvent this issue, we propose a novel formulation of contrastive learning using semantic similarity between instances called Similarity Contrastive Estimation (SCE). Our training objective is a soft contrastive one that brings the positives closer and estimates a continuous distribution to push or pull negative instances based on their learned similarities. We validate empirically our approach on both image and video representation learning. We show that SCE performs competitively with the state of the art on the ImageNet linear evaluation protocol for fewer pretraining epochs and that it generalizes to several downstream image tasks. We also show that SCE reaches state-of-the-art results for pretraining video representation and that the learned representation can generalize to video downstream tasks.

对比表现学习已被证明是一种有效的自我监督学习方法。大多数成功的方法都是基于噪声对比估计（NCE）范式，并将实例视图的视图视为阳性和其他情况，作为阳性应与其对比的噪声。但是，数据集中的所有实例都是从相同的分布和共享底层语义信息中汲取，这些语义信息不应被视为噪声。我们认为，良好的数据表示包含实例之间的关系或语义相似性。对比学习隐含地学习关系，但认为负面的噪音是对学习关系质量有害的噪音，因此是象征性的质量。为了规避这个问题，我们提出了一种使用称为相似性对比估计（SCE）之间的情况之间的语义相似性的对比学习的新颖性。我们的培训目标可以被视为柔和的对比学习。我们提出了持续分配以基于其语义相似性推动或拉动实例的持续分配。目标相似性分布从弱增强的情况计算并锐化以消除无关的关系。每个弱增强实例都与一个强大的增强实例配对，该实例对比其积极的同时保持目标相似性分布。实验结果表明，我们所提出的SCE在各种数据集中优于其基线MoCov2和RESSL，并对ImageNet线性评估协议上的最先进的算法具有竞争力。

最近无监督的表示学习方法已经通过学习表示不变的数据增强，例如随机裁剪和彩色抖动等数据增强来生效。然而，如果依赖于数据增强的特征，例如，位置或色敏，则这种不变性可能对下游任务有害。这不是一个不监督学习的问题;我们发现即使在监督学习中也会发生这种情况，因为它还学会预测实例所有增强样本的相同标签。为避免此类失败并获得更广泛的表示，我们建议优化辅助自我监督损失，创建的AGESELF，了解两个随机增强样本之间的增强参数（例如，裁剪位置，颜色调整强度）的差异。我们的直觉是，Augelf鼓励在学习的陈述中保留增强信息，这可能有利于其可转让性。此外，Augself可以很容易地纳入最近的最先进的表示学习方法，其额外的培训成本可忽略不计。广泛的实验表明，我们的简单想法一直在各种转移学习情景中始终如一地提高了由监督和无监督方法所学到的表示的可转移性。代码可在https://github.com/hankook/augsfir。

自我监督学习的最新进展证明了多种视觉任务的有希望的结果。高性能自我监督方法中的一个重要成分是通过培训模型使用数据增强，以便在嵌入空间附近的相同图像的不同增强视图。然而，常用的增强管道整体地对待图像，忽略图像的部分的语义相关性-e.g。主题与背景 - 这可能导致学习杂散相关性。我们的工作通过调查一类简单但高度有效的“背景增强”来解决这个问题，这鼓励模型专注于语义相关内容，劝阻它们专注于图像背景。通过系统的调查，我们表明背景增强导致在各种任务中跨越一系列最先进的自我监督方法（MOCO-V2，BYOL，SWAV）的性能大量改进。 $ \ SIM $ + 1-2％的ImageNet收益，使得与监督基准的表现有关。此外，我们发现有限标签设置的改进甚至更大（高达4.2％）。背景技术增强还改善了许多分布换档的鲁棒性，包括天然对抗性实例，想象群-9，对抗性攻击，想象成型。我们还在产生了用于背景增强的显着掩模的过程中完全无监督的显着性检测进展。

Contrastive learning has become a key component of self-supervised learning approaches for computer vision. By learning to embed two augmented versions of the same image close to each other and to push the embeddings of different images apart, one can train highly transferable visual representations. As revealed by recent studies, heavy data augmentation and large sets of negatives are both crucial in learning such representations. At the same time, data mixing strategies, either at the image or the feature level, improve both supervised and semi-supervised learning by synthesizing novel examples, forcing networks to learn more robust features. In this paper, we argue that an important aspect of contrastive learning, i.e. the effect of hard negatives, has so far been neglected. To get more meaningful negative samples, current top contrastive self-supervised learning approaches either substantially increase the batch sizes, or keep very large memory banks; increasing memory requirements, however, leads to diminishing returns in terms of performance. We therefore start by delving deeper into a top-performing framework and show evidence that harder negatives are needed to facilitate better and faster learning. Based on these observations, and motivated by the success of data mixing, we propose hard negative mixing strategies at the feature level, that can be computed on-the-fly with a minimal computational overhead. We exhaustively ablate our approach on linear classification, object detection, and instance segmentation and show that employing our hard negative mixing procedure improves the quality of visual representations learned by a state-of-the-art self-supervised learning method.Project page: https://europe.naverlabs.com/mochi 34th Conference on Neural Information Processing Systems (NeurIPS 2020),