事实证明,知识蒸馏是使用教师模型的预测来改善学生模型的一项有效技术。但是,最近的工作表明,在数据中的亚组中,平均效率的提高并不统一,尤其是在稀有亚组和类别上的准确性通常可能以准确性为代价。为了在可能遵循长尾分配的课程中保持强劲的表现,我们开发了蒸馏技术,这些技术是为了改善学生最差的级别表现而定制的。具体来说,我们为教师和学生介绍了不同组合的强大优化目标,并进一步允许在整体准确性和强大的最差目标之间进行任何权衡训练。我们从经验上表明,与其他基线方法相比,我们强大的蒸馏技术不仅可以实现更好的最差级别性能,而且还可以改善整体性能和最差的级别性能之间的权衡。从理论上讲,我们提供有关在目标培训健壮学生时使一名好老师的见解。
translated by 谷歌翻译
虽然神经网络在平均病例的性能方面对分类任务的成功显着,但它们通常无法在某些数据组上表现良好。这样的组信息可能是昂贵的;因此,即使在培训数据不可用的组标签不可用,较稳健性和公平的最新作品也提出了改善最差组性能的方法。然而,这些方法通常在培训时间使用集团信息的表现不佳。在这项工作中,我们假设没有组标签的较大数据集一起访问少量组标签。我们提出了一个简单的两步框架,利用这个部分组信息来提高最差组性能:训练模型以预测训练数据的丢失组标签,然后在强大的优化目标中使用这些预测的组标签。从理论上讲,我们在最差的组性能方面为我们的方法提供泛化界限,展示了泛化误差如何相对于培训点总数和具有组标签的培训点的数量。凭经验,我们的方法优于不使用群组信息的基线表达,即使只有1-33%的积分都有组标签。我们提供消融研究,以支持我们框架的稳健性和可扩展性。
translated by 谷歌翻译
Real-world classification problems typically exhibit an imbalanced or long-tailed label distribution, wherein many labels are associated with only a few samples. This poses a challenge for generalisation on such labels, and also makes naïve learning biased towards dominant labels. In this paper, we present two simple modifications of standard softmax cross-entropy training to cope with these challenges. Our techniques revisit the classic idea of logit adjustment based on the label frequencies, either applied post-hoc to a trained model, or enforced in the loss during training. Such adjustment encourages a large relative margin between logits of rare versus dominant labels. These techniques unify and generalise several recent proposals in the literature, while possessing firmer statistical grounding and empirical performance. A reference implementation of our methods is available at: https://github.com/google-research/google-research/tree/master/logit_adjustment.Recently, long-tail learning has received renewed interest in the context of neural networks. Two active strands of work involve post-hoc normalisation of the classification weights [
translated by 谷歌翻译
现代机器学习问题中的不平衡数据集是司空见惯的。具有敏感属性的代表性课程或群体的存在导致关于泛化和公平性的担忧。这种担忧进一步加剧了大容量深网络可以完全适合培训数据,似乎在训练期间达到完美的准确性和公平,但在测试期间表现不佳。为了解决这些挑战,我们提出了自动化,一个自动设计培训损失功能的双层优化框架,以优化准确性和寻求公平目标的混合。具体地,较低级别的问题列举了模型权重,并且上级问题通过监视和优化通过验证数据的期望目标来调谐损耗功能。我们的损耗设计通过采用参数跨熵损失和个性化数据增强方案,可以为类/组进行个性化处理。我们评估我们对不平衡和群体敏感分类的应用方案的方法的好处和性能。广泛的经验评估表明了自动矛盾最先进的方法的益处。我们的实验结果与损耗功能设计的理论见解和培训验证分裂的好处相辅相成。所有代码都是可用的开源。
translated by 谷歌翻译
重要性加权是一种处理分销班次的经典技术。然而,事先工作呈现出强大的实证和理论证据,证明重要性重量对过度分辨的神经网络没有影响。重要性加权与过度分辨率的神经网络的培训真正不相容吗?我们的论文在负面回答。我们表明重要的权重不是因为过度分辨率,而是因为使用像物流或交叉熵损失等指数尾损失。作为一种补救措施,我们表明多项式尾损失恢复了重要性重量在校正过度分配模型中的分布换档的影响。我们表征了梯度下降的行为,其具有过度分辨的线性模型的重要性加权多项式损耗,并且理论上证明了在标签换档设置中使用多环尾损失的优点。令人惊讶的是,我们的理论表明,使用通过以指数来引入经典无偏的重要性重量而获得的权重可以提高性能。最后,我们展示了我们对亚潜班班和标签移位数据集的神经网络实验的分析的实际价值。重新重复时,我们的损耗函数可以在测试精度的高达9%的跨熵优先于重复的交叉熵。我们的损耗功能还提供了与校正分配换档的最先进的方法可比或甚至超过的测试精度。
translated by 谷歌翻译
除了使用硬标签的标准监督学习外,通常在许多监督学习设置中使用辅助损失来改善模型的概括。例如,知识蒸馏增加了第二个教师模仿模型训练的损失,在该培训中,教师可能是一个验证的模型,可以输出比标签更丰富的分布。同样,在标记数据有限的设置中,弱标记信息以标签函数的形式使用。此处引入辅助损失来对抗标签函数,这些功能可能是基于嘈杂的规则的真实标签近似值。我们解决了学习以原则性方式结合这些损失的问题。我们介绍AMAL,该AMAL使用元学习在验证度量上学习实例特定的权重,以实现损失的最佳混合。在许多知识蒸馏和规则降解域中进行的实验表明,Amal在这些领域中对竞争基准的增长可显着。我们通过经验分析我们的方法,并分享有关其提供性能提升的机制的见解。
translated by 谷歌翻译
近年来,已取得了巨大进展,以通过半监督学习(SSL)来纳入未标记的数据来克服效率低下的监督问题。大多数最先进的模型是基于对未标记的数据追求一致的模型预测的想法,该模型被称为输入噪声,这称为一致性正则化。尽管如此,对其成功的原因缺乏理论上的见解。为了弥合理论和实际结果之间的差距,我们在本文中提出了SSL的最坏情况一致性正则化技术。具体而言,我们首先提出了针对SSL的概括,该概括由分别在标记和未标记的训练数据上观察到的经验损失项组成。在这种界限的激励下,我们得出了一个SSL目标,该目标可最大程度地减少原始未标记的样本与其多重增强变体之间最大的不一致性。然后,我们提供了一种简单但有效的算法来解决提出的最小问题,从理论上证明它会收敛到固定点。五个流行基准数据集的实验验证了我们提出的方法的有效性。
translated by 谷歌翻译
标签 - 不平衡和组敏感分类中的目标是优化相关的指标,例如平衡错误和相同的机会。经典方法,例如加权交叉熵,在训练深网络到训练(TPT)的终端阶段时,这是超越零训练误差的训练。这种观察发生了最近在促进少数群体更大边值的直观机制之后开发启发式替代品的动力。与之前的启发式相比,我们遵循原则性分析,说明不同的损失调整如何影响边距。首先,我们证明,对于在TPT中训练的所有线性分类器,有必要引入乘法,而不是添加性的Logit调整,以便对杂项边缘进行适当的变化。为了表明这一点,我们发现将乘法CE修改的连接到成本敏感的支持向量机。也许是违反,我们还发现,在培训开始时,相同的乘法权重实际上可以损害少数群体。因此,虽然在TPT中,添加剂调整无效,但我们表明它们可以通过对乘法重量的初始负效应进行抗衡来加速会聚。通过这些发现的动机,我们制定了矢量缩放(VS)丢失,即捕获现有技术作为特殊情况。此外,我们引入了对群体敏感分类的VS损失的自然延伸,从而以统一的方式处理两种常见类型的不平衡(标签/组)。重要的是,我们对最先进的数据集的实验与我们的理论见解完全一致,并确认了我们算法的卓越性能。最后,对于不平衡的高斯 - 混合数据,我们执行泛化分析,揭示平衡/标准错误和相同机会之间的权衡。
translated by 谷歌翻译
学习算法的目标之一是补充和减轻人类决策者的负担。算法可以自行预测的专家延期设置,也可以将决定推迟到下游专家有助于实现这一目标。这种环境的一个基本方面是需要学习改善人类弱点的互补预测因子,而不是学习预测因素以优化平均错误。在这项工作中,我们提供了对专家延期中学习补充预测指标的好处的第一个理论分析。为了有效地学习此类预测因素,我们考虑了一个始终如一的替代损失功能的家族,以延期专家并分析其理论特性。最后,我们设计的主动学习方案需要最少的人类专家预测数据,以学习准确的延期系统。
translated by 谷歌翻译
Deep learning algorithms can fare poorly when the training dataset suffers from heavy class-imbalance but the testing criterion requires good generalization on less frequent classes. We design two novel methods to improve performance in such scenarios. First, we propose a theoretically-principled label-distribution-aware margin (LDAM) loss motivated by minimizing a margin-based generalization bound. This loss replaces the standard cross-entropy objective during training and can be applied with prior strategies for training with class-imbalance such as re-weighting or re-sampling. Second, we propose a simple, yet effective, training schedule that defers re-weighting until after the initial stage, allowing the model to learn an initial representation while avoiding some of the complications associated with re-weighting or re-sampling. We test our methods on several benchmark vision tasks including the real-world imbalanced dataset iNaturalist 2018. Our experiments show that either of these methods alone can already improve over existing techniques and their combination achieves even better performance gains 1 .
translated by 谷歌翻译
尽管自我监督学习(SSL)方法取得了经验成功,但尚不清楚其表示的哪些特征导致了高下游精度。在这项工作中,我们表征了SSL表示应该满足的属性。具体而言,我们证明了必要和充分的条件,因此,对于给出的数据增强的任何任务,在该表示形式上训练的所需探针(例如,线性或MLP)具有完美的准确性。这些要求导致一个统一的概念框架,用于改善现有的SSL方法并得出新方法。对于对比度学习,我们的框架规定了对以前的方法(例如使用不对称投影头)的简单但重大改进。对于非对比度学习,我们使用框架来得出一个简单新颖的目标。我们所得的SSL算法在标准基准测试上的表现优于基线,包括Imagenet线性探测的SHAV+多螺旋桨。
translated by 谷歌翻译
In this paper, we present a simple yet effective method (ABSGD) for addressing the data imbalance issue in deep learning. Our method is a simple modification to momentum SGD where we leverage an attentional mechanism to assign an individual importance weight to each gradient in the mini-batch. Unlike many existing heuristic-driven methods for tackling data imbalance, our method is grounded in {\it theoretically justified distributionally robust optimization (DRO)}, which is guaranteed to converge to a stationary point of an information-regularized DRO problem. The individual-level weight of a sampled data is systematically proportional to the exponential of a scaled loss value of the data, where the scaling factor is interpreted as the regularization parameter in the framework of information-regularized DRO. Compared with existing class-level weighting schemes, our method can capture the diversity between individual examples within each class. Compared with existing individual-level weighting methods using meta-learning that require three backward propagations for computing mini-batch stochastic gradients, our method is more efficient with only one backward propagation at each iteration as in standard deep learning methods. To balance between the learning of feature extraction layers and the learning of the classifier layer, we employ a two-stage method that uses SGD for pretraining followed by ABSGD for learning a robust classifier and finetuning lower layers. Our empirical studies on several benchmark datasets demonstrate the effectiveness of the proposed method.
translated by 谷歌翻译
深度神经网络的成功在很大程度上取决于大量高质量注释的数据的可用性,但是这些数据很难或昂贵。由此产生的标签可能是类别不平衡,嘈杂或人类偏见。从不完美注释的数据集中学习无偏分类模型是一项挑战,我们通常会遭受过度拟合或不足的折磨。在这项工作中,我们彻底研究了流行的软马克斯损失和基于保证金的损失,并提供了一种可行的方法来加强通过最大化最小样本余量来限制的概括误差。我们为此目的进一步得出了最佳条件,该条件指示了类原型应锚定的方式。通过理论分析的激励,我们提出了一种简单但有效的方法,即原型锚定学习(PAL),可以轻松地将其纳入各种基于学习的分类方案中以处理不完美的注释。我们通过对合成和现实世界数据集进行广泛的实验来验证PAL对班级不平衡学习和降低噪声学习的有效性。
translated by 谷歌翻译
深入学习在现代分类任务中取得了许多突破。已经提出了众多架构用于不同的数据结构,但是当涉及丢失功能时,跨熵损失是主要的选择。最近,若干替代损失已经看到了深度分类器的恢复利益。特别是,经验证据似乎促进了方形损失,但仍然缺乏理论效果。在这项工作中,我们通过系统地研究了在神经切线内核(NTK)制度中的过度分化的神经网络的表现方式来促进对分类方面损失的理论理解。揭示了关于泛化误差,鲁棒性和校准错误的有趣特性。根据课程是否可分离,我们考虑两种情况。在一般的不可分类案例中,为错误分类率和校准误差建立快速收敛速率。当类是可分离的时,错误分类率改善了速度快。此外,经过证明得到的余量被证明是低于零的较低,提供了鲁棒性的理论保证。我们希望我们的调查结果超出NTK制度并转化为实际设置。为此,我们对实际神经网络进行广泛的实证研究,展示了合成低维数据和真实图像数据中方损的有效性。与跨熵相比,方形损耗具有可比的概括误差,但具有明显的鲁棒性和模型校准的优点。
translated by 谷歌翻译
我们提出了简单的主动采样和重新重量策略,以优化最小最大公平性,可以应用于通过损耗最小化学习的任何分类或回归模型。我们的方法背后的关键直觉是在每个TIMESTEP中使用来自当前模型中最差的组的DataPoint,以更新模型。实施的易于实现和我们稳健的制定的一般性使其成为提高糟糕表现群体的模型性能的有吸引力的选择。对于凸起的学习问题,如线性或逻辑回归,我们提供了对我们的策略的细粒度分析,证明了其收敛速度对Min-Max Fair解决方案。
translated by 谷歌翻译
知识蒸馏是一种培训小型学生网络的流行技术,以模仿更大的教师模型,例如网络的集合。我们表明,虽然知识蒸馏可以改善学生泛化,但它通常不得如此普遍地工作:虽然在教师和学生的预测分布之间,甚至在学生容量的情况下,通常仍然存在令人惊讶的差异完美地匹配老师。我们认为优化的困难是为什么学生无法与老师匹配的关键原因。我们还展示了用于蒸馏的数据集的细节如何在学生与老师匹配的紧密关系中发挥作用 - 以及教师矛盾的教师并不总是导致更好的学生泛化。
translated by 谷歌翻译
如何培训理想的老师进行知识蒸馏仍然是一个悬而未决的问题。人们普遍观察到,将教师最小化经验风险不一定会产生表现最好的学生,这表明教师网络培训中的共同实践与蒸馏目标之间的基本差异。为了填补这一空白,我们提出了一个新颖的以学生为导向的教师网络培训框架Soteacher,这是受到最新发现的启发,即学生的表现取决于教师近似培训样本的真正标签分布的能力。从理论上讲,我们确定(1)具有适当评分规则的经验风险最小化器,如果假设函数是局部lipschitz在训练样本周围连续的,则可以证明训练数据的真实标签分布; (2)当使用数据扩展进行培训时,需要一个额外的约束,使最小化器必须在同一培训输入的增强视图中产生一致的预测。鉴于我们的理论,Soteacher通过结合Lipschitz正则化和​​一致性正则化来翻新经验风险最小化。值得一提的是,Soteacher几乎适用于所有教师学生的建筑对,在教师的培训时不需要对学生的先验知识,并且几乎没有任何计算开销。两个基准数据集的实验证实,Soteacher可以在各种知识蒸馏算法和教师成对的各种知识蒸馏算法中显着和一致地提高学生的绩效。
translated by 谷歌翻译
Learned classifiers should often possess certain invariance properties meant to encourage fairness, robustness, or out-of-distribution generalization. However, multiple recent works empirically demonstrate that common invariance-inducing regularizers are ineffective in the over-parameterized regime, in which classifiers perfectly fit (i.e. interpolate) the training data. This suggests that the phenomenon of ``benign overfitting," in which models generalize well despite interpolating, might not favorably extend to settings in which robustness or fairness are desirable. In this work we provide a theoretical justification for these observations. We prove that -- even in the simplest of settings -- any interpolating learning rule (with arbitrarily small margin) will not satisfy these invariance properties. We then propose and analyze an algorithm that -- in the same setting -- successfully learns a non-interpolating classifier that is provably invariant. We validate our theoretical observations on simulated data and the Waterbirds dataset.
translated by 谷歌翻译
We introduce a tunable loss function called $\alpha$-loss, parameterized by $\alpha \in (0,\infty]$, which interpolates between the exponential loss ($\alpha = 1/2$), the log-loss ($\alpha = 1$), and the 0-1 loss ($\alpha = \infty$), for the machine learning setting of classification. Theoretically, we illustrate a fundamental connection between $\alpha$-loss and Arimoto conditional entropy, verify the classification-calibration of $\alpha$-loss in order to demonstrate asymptotic optimality via Rademacher complexity generalization techniques, and build-upon a notion called strictly local quasi-convexity in order to quantitatively characterize the optimization landscape of $\alpha$-loss. Practically, we perform class imbalance, robustness, and classification experiments on benchmark image datasets using convolutional-neural-networks. Our main practical conclusion is that certain tasks may benefit from tuning $\alpha$-loss away from log-loss ($\alpha = 1$), and to this end we provide simple heuristics for the practitioner. In particular, navigating the $\alpha$ hyperparameter can readily provide superior model robustness to label flips ($\alpha > 1$) and sensitivity to imbalanced classes ($\alpha < 1$).
translated by 谷歌翻译
A key learning scenario in large-scale applications is that of federated learning, where a centralized model is trained based on data originating from a large number of clients. We argue that, with the existing training and inference, federated models can be biased towards different clients. Instead, we propose a new framework of agnostic federated learning, where the centralized model is optimized for any target distribution formed by a mixture of the client distributions. We further show that this framework naturally yields a notion of fairness. We present data-dependent Rademacher complexity guarantees for learning with this objective, which guide the definition of an algorithm for agnostic federated learning. We also give a fast stochastic optimization algorithm for solving the corresponding optimization problem, for which we prove convergence bounds, assuming a convex loss function and hypothesis set. We further empirically demonstrate the benefits of our approach in several datasets. Beyond federated learning, our framework and algorithm can be of interest to other learning scenarios such as cloud computing, domain adaptation, drifting, and other contexts where the training and test distributions do not coincide. MotivationA key learning scenario in large-scale applications is that of federated learning. In that scenario, a centralized model is trained based on data originating from a large number of clients, which may be mobile phones, other mobile devices, or sensors (Konečnỳ, McMahan, Yu, Richtárik, Suresh, and Bacon, 2016b;Konečnỳ, McMahan, Ramage, and Richtárik, 2016a). The training data typically remains distributed over the clients, each with possibly unreliable or relatively slow network connections.Federated learning raises several types of issues and has been the topic of multiple research efforts. These include systems, networking and communication bottleneck problems due to frequent exchanges between the central server and the clients . To deal with such problems, suggested an averaging technique that consists of transmitting the central model to a subset of clients, training it with the data locally available, and averaging the local updates. Smith et al. (2017) proposed to further leverage the relationship between clients, assumed to be known, and cast
translated by 谷歌翻译