Class imbalance is a common challenge in machine learning and data mining, often leading to suboptimal performance in classifiers. While deep learning excels in feature extraction, its performance still deteriorates under imbalanced data. In this work, we propose a novel activation function, named OGAB, designed to alleviate class imbalance in deep learning classifiers. OGAB incorporates orthogonality and group-aware bias learning to enhance feature distinguishability in imbalanced scenarios without explicitly requiring label information. Our key insight is that activation functions can be used to introduce strong inductive biases that can address complex data challenges beyond traditional non-linearity. Our work demonstrates that orthogonal transformations can preserve information about minority classes by maintaining feature independence, thereby preventing the dominance of majority classes in the embedding space. Further, the proposed group-aware bias mechanism automatically identifies data clusters and adjusts embeddings to enhance class separability without the need for explicit supervision. Unlike existing approaches that address class imbalance through preprocessing data modifications or post-processing corrections, our proposed approach tackles class imbalance during the training phase at the embedding learning level, enabling direct integration with the learning process. We demonstrate the effectiveness of our solution on both real-world and synthetic imbalanced datasets, showing consistent performance improvements over both traditional and learnable activation functions.
Adaptive Parametric Activation: Unifying and Generalising Activation Functions Across Tasks
Huawei Noah’s Ark Lab（华为诺亚实验室）；Department of Computing, Imperial College London（帝国理工学院计算机系）；Department of Computer Science, University of Liverpool（利物浦大学计算机科学系）；Department of Engineering, King’s College London（伦敦国王学院工程系）
Purdue University（普渡大学）；The University of Chicago（芝加哥大学）；National University of Singapore（新加坡国立大学）
Overfitting frequently occurs in deep learning. In this paper, we propose a novel regularization method called Drop-Activation to reduce overfitting and improve generalization. The key idea is to drop nonlinear activation functions by setting them to be identity functions randomly during training time. During testing, we use a deterministic network with a new activation function to encode the average effect of dropping activations randomly. Our theoretical analyses support the regularization effect of Drop-Activation as implicit parameter reduction and verify its capability to be used together…
Implicitly defined, continuous, differentiable signal representations parameterized by neural networks have emerged as a powerful paradigm, offering many possible benefits over conventional representations. However, current network architectures for such implicit neural representations are incapable of modeling signals with fine detail, and fail to represent a signal's spatial and temporal derivatives, despite the fact that these are essential to many physical signals defined implicitly as the solution to partial differential equations. We propose to leverage periodic activation functions for …
We propose a novel output layer activation function, which we name ASTra (Asymmetric Sigmoid Transfer function), which makes the classification of minority examples, in scenarios of high imbalance, more tractable. We combine this with a loss function that helps to effectively target minority misclassification. These two methods can be used together or separately, with their combination recommended for the most severely imbalanced cases. The proposed approach is tested on datasets with IRs from 588.24 to 4000 and very few minority examples (in some datasets, as few as five). Results using neura…
Stanford University（斯坦福大学）；Purdue University（普渡大学）；Institute for Advanced Study（高等研究院）
We consider networks, trained via stochastic gradient descent to minimize $\ell_2$ loss, with the training labels perturbed by independent noise at each iteration. We characterize the behavior of the training dynamics near any parameter vector that achieves zero training error, in terms of an implicit regularization term corresponding to the sum over the data points, of the squared $\ell_2$ norm of the gradient of the model with respect to the parameter vector, evaluated at each data point. This holds for networks of any connectivity, width, depth, and choice of activation function. We interpr…
Carnegie Mellon University（卡内基梅隆大学）；Bosch Center for AI（博世人工智能中心）
Neural networks with sinusoidal activations have been proposed as an alternative to networks with traditional activation functions. Despite their promise, particularly for learning implicit models, their training behavior is not yet fully understood, leading to a number of empirical design choices that are not well justified. In this work, we first propose a simplified version of such sinusoidal neural networks, which allows both for easier practical implementation and simpler theoretical analysis. We then analyze the behavior of these networks from the neural tangent kernel perspective and de…
University of British Columbia（不列颠哥伦比亚大学）；University of California, Santa Barbara（加州大学圣巴巴拉分校）
Various logit-adjusted parameterizations of the cross-entropy (CE) loss have been proposed as alternatives to weighted CE for training large models on label-imbalanced data far beyond the zero train error regime. The driving force behind those designs has been the theory of implicit bias, which for linear(ized) models, explains why they successfully induce bias on the optimization path towards solutions that favor minorities. Aiming to extend this theory to non-linear models, we investigate the implicit geometry of classifiers and embeddings that are learned by different CE parameterizations. …
Reducing the precision of weights and activation functions in neural network training, with minimal impact on performance, is essential for the deployment of these models in resource-constrained environments. We apply mean-field techniques to networks with quantized activations in order to evaluate the degree to which quantization degrades signal propagation at initialization. We derive initialization schemes which maximize signal propagation in such networks and suggest why this is helpful for generalization. Building on these results, we obtain a closed form implicit equation for $L_{\max}$,…
Tsinghua University（清华大学）；University of Illinois at Chicago（伊利诺伊大学芝加哥分校）
Learning to hash has been widely applied to approximate nearest neighbor search for large-scale multimedia retrieval, due to its computation efficiency and retrieval quality. Deep learning to hash, which improves retrieval quality by end-to-end representation learning and hash encoding, has received increasing attention recently. Subject to the ill-posed gradient difficulty in the optimization with sign activations, existing deep learning to hash methods need to first learn continuous representations and then generate binary hash codes in a separated binarization step, which suffer from substa…
Towards the Connection between Activation Sparsity and Flat Minima
State Key Laboratory for Novel Software Technology, Nanjing University（南京大学新型软件技术国家重点实验室）；Institute of Brain-Machine Interface, Nanjing University（南京大学脑机接口研究院）；School of Computer Science and Engineering, Southeast University（东南大学计算机科学与工程学院）
ARC Centre of Excellence for Robotic Vision（ARC机器人视觉卓越中心）；University of Adelaide（阿德莱德大学）
Deep feedforward neural networks with piecewise linear activations are currently producing the state-of-the-art results in several public datasets. The combination of deep learning models and piecewise linear activation functions allows for the estimation of exponentially complex functions with the use of a large number of subnetworks specialized in the classification of similar input examples. During the training process, these subnetworks avoid overfitting with an implicit regularization scheme based on the fact that they must share their parameters with other subnetworks. Using this framewo…
In this work, we evaluate how neural networks with periodic activation functions can be leveraged to reliably compress large multidimensional medical image datasets, with proof-of-concept application to 4D diffusion-weighted MRI (dMRI). In the medical imaging landscape, multidimensional MRI is a key area of research for developing biomarkers that are both sensitive and specific to the underlying tissue microstructure. However, the high-dimensional nature of these data poses a challenge in terms of both storage and sharing capabilities and associated costs, requiring appropriate algorithms able…
Indian Institute of Technology Madras（印度理工学院马德拉斯分校）
In the era of Deep Neural Network based solutions for a variety of real-life tasks, having a compact and energy-efficient deployable model has become fairly important. Most of the existing deep architectures use Rectifier Linear Unit (ReLU) activation. In this paper, we propose a novel idea of rotating the ReLU activation to give one more degree of freedom to the architecture. We show that this activation wherein the rotation is learned via training results in the elimination of those parameters/filters in the network which are not important for the task. In other words, rotated ReLU seems to …
Robustness of machine learning models on ever-changing real-world data is critical, especially for applications affecting human well-being such as content moderation. New kinds of abusive language continually emerge in online discussions in response to current events (e.g., COVID-19), and the deployed abuse detection systems should be updated regularly to remain accurate. In this paper, we show that general abusive language classifiers tend to be fairly reliable in detecting out-of-domain explicitly abusive utterances but fail to detect new types of more subtle, implicit abuse. Next, we propos…
Node classification is an important task to solve in graph-based learning. Even though a lot of work has been done in this field, imbalance is neglected. Real-world data is not perfect, and is imbalanced in representations most of the times. Apart from text and images, data can be represented using graphs, and thus addressing the imbalance in graphs has become of paramount importance. In the context of node classification, one class has less examples than others. Changing data composition is a popular way to address the imbalance in node classification. This is done by resampling the data to b…
All-around Neural Collapse for Imbalanced Classification
MIIT Key Laboratory of Pattern Analysis and Machine Intelligence（MIIT模式分析与智能关键实验室）；College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics (NUAA)（南京航空航天大学计算机科学与技术学院）
Imagia（Imagia公司）；Dalhousie University（达尔豪斯大学）；Mila, Université de Montréal（蒙特利尔大学米拉研究所）；Polish Academy of Sciences（波兰科学院）
We present an approach for unsupervised domain adaptation---with a strong focus on practical considerations of within-domain class imbalance and between-domain class distribution shift---from a class-conditioned domain alignment perspective. Current methods for class-conditioned domain alignment aim to explicitly minimize a loss function based on pseudo-label estimations of the target domain. However, these methods suffer from pseudo-label bias in the form of error accumulation. We propose a method that removes the need for explicit optimization of model parameters from pseudo-labels directly.…
Brown University（布朗大学）；Massachusetts Institute of Technology（麻省理工学院）；Pacific Northwest National Laboratory（太平洋西北国家实验室）
We propose two approaches of locally adaptive activation functions namely, layer-wise and neuron-wise locally adaptive activation functions, which improve the performance of deep and physics-informed neural networks. The local adaptation of activation function is achieved by introducing a scalable parameter in each layer (layer-wise) and for every neuron (neuron-wise) separately, and then optimizing it using a variant of stochastic gradient descent algorithm. In order to further increase the training speed, an activation slope based slope recovery term is added in the loss function, which furt…
Tilburg University（蒂尔堡大学）；Eindhoven University of Technology（埃因霍温理工大学）；Hasselt University（哈瑟尔特大学）；Central University of Las Villas（拉斯维亚斯中央大学）；São Paulo State University（圣保罗州立大学）
This paper presents a Fuzzy Cognitive Map model to quantify implicit bias in structured datasets where features can be numeric or discrete. In our proposal, problem features are mapped to neural concepts that are initially activated by experts when running what-if simulations, whereas weights connecting the neural concepts represent absolute correlation/association patterns between features. In addition, we introduce a new reasoning mechanism equipped with a normalization-like transfer function that prevents neurons from saturating. Another advantage of this new reasoning mechanism is that it …
University of Pittsburgh（匹兹堡大学）；Robert Bosch LLC（罗伯特·博世有限公司）
We propose thresholding as an approach to deal with class imbalance. We define the concept of thresholding as a process of determining a decision boundary in the presence of a tunable parameter. The threshold is the maximum value of this tunable parameter where the conditions of a certain decision are satisfied. We show that thresholding is applicable not only for linear classifiers but also for non-linear classifiers. We show that this is the implicit assumption for many approaches to deal with class imbalance in linear classifiers. We then extend this paradigm beyond linear classification an…
Heriot-Watt University（赫瑞-瓦特大学）；University of Edinburgh（爱丁堡大学）；SeeByte Ltd.（SeeByte有限公司）
Meta-Learning (ML) has proven to be a useful tool for training Few-Shot Learning (FSL) algorithms by exposure to batches of tasks sampled from a meta-dataset. However, the standard training procedure overlooks the dynamic nature of the real-world where object classes are likely to occur at different frequencies. While it is generally understood that imbalanced tasks harm the performance of supervised methods, there is no significant research examining the impact of imbalanced meta-datasets on the FSL evaluation task. This study exposes the magnitude and extent of this problem. Our results show…
Neural networks trained via gradient descent with random initialization and without any regularization enjoy good generalization performance in practice despite being highly overparametrized. A promising direction to explain this phenomenon is to study how initialization and overparametrization affect convergence and implicit bias of training algorithms. In this paper, we present a novel analysis of single-hidden-layer linear networks trained under gradient flow, which connects initialization, optimization, and overparametrization. Firstly, we show that the squared loss converges exponentially…
Developing Explainable Machine Learning Model using Augmented Concept Activation Vector
Groningen University（格罗宁根大学）；Konya Food and Agriculture University（科尼亚食品与农业大学）；Rotterdam University（鹿特丹大学）；Research Center Creating 010（010创新研究中心）；Research and Data Center, Ministry of Justice and Security, The Netherlands（荷兰司法与安全部研究与数据中心）
ETH Z\"urich, D-Math, R\"amistrasse 101, CH-8092 Z\"urich, Switzerland
Randomized neural networks (randomized NNs), where only the terminal layer's weights are optimized constitute a powerful model class to reduce computational time in training the neural network model. At the same time, these models generalize surprisingly well in various regression and classification tasks. In this paper, we give an exact macroscopic characterization (i.e., a characterization in function space) of the generalization behavior of randomized, shallow NNs with ReLU activation (RSNs). We show that RSNs correspond to a generalized additive model (GAM)-typed regression in which infini…
