A network architecture improves classification performance on various datasets.
problem Improving classification performance on diverse datasets.
method Divide fully connected layer into three levels, learn existing layers, cluster similar classes, reclassify using clustering masks.
result Achieved state-of-the-art performance with an error rate of 11.56% on Cifar-100.
Proposes LML layer for multi-label predictions with k labels.
problem Efficient multi-label prediction with limited labels.
method Probabilistic multi-label modeling, efficient forward and backward passes.
result Improves top-k recall and accuracy in multi-label tasks.
Study improves top-k set prediction with low cardinality.
problem Improving top-k set prediction accuracy with low cardinality.
method Introduces new target loss function and surrogate losses.
result Demonstrates effectiveness of cardinality-aware algorithms.
Improved sample efficiency with normalized RBF kernels in neural networks.
problem Learning more with less data in deep learning models.
method Two-phase method to train neural networks with normalized RBF kernels as output layer.
result Normalized RBF kernel networks achieve higher sample efficiency, compactness, and separability.
Improved VGG networks enhance image classification accuracy.
problem Enhancing image classification accuracy using modified VGG architectures.
method Two improved VGG architectures were created by freezing the first two blocks and applying different dilation rates in the last three blocks.
result Significant out-performance on image classification tasks on CIFAR-10 and CIFAR-100 datasets.
Unified view of contrastive and supervised learning improves model performance.
problem Improving model performance in classification, robustness, and detection.
method Hybrid discriminative-generative training of energy-based models.
result Improved performance on classification, robustness, out-of-distribution detection, and calibration.
Study linear transformations' effects on data augmentation for improved estimation.
problem Improving performance in image and text classification tasks.
method Examined a family of linear transformations in over-parametrized linear regression settings.
result Transformations that preserve labels or mix data can improve estimation.
Convolutional DKMs improve kernel methods on MNIST, CIFAR-10, and CIFAR-100.
problem Improving kernel methods for image classification.
method Developed a novel inter-domain inducing point approximation and introduced various techniques to extend DKMs to convolutional networks.
result Achieved state-of-the-art performance on image classification benchmarks.
This research tackles image classification with noise, proposing committees of CNNs.
problem Image classification with concurrent feature and label noise.
method Committees of Convolutional Neural Networks (CNNs) for MNIST, CIFAR-10, and CIFAR-100 datasets.
result Committees outperform single models in noisy conditions, especially on difficult datasets.
A new method uses optimal transport for semi-supervised classification.
problem Semi-supervised learning with limited labeled data.
method Optimal transport formulation with Sinkhorn iteration for label assignment.
result Improved performance on CIFAR-10, CIFAR-100, and SVHN datasets compared to FixMatch.
This paper examines how to calibrate ensemble members for better prediction accuracy.
problem Improper calibration of deep neural networks leads to unreliable probability estimates.
method Theoretical analysis and empirical evaluation on CIFAR-100 dataset.
result Well-calibrated ensemble members do not guarantee a well-calibrated ensemble prediction, but a well-calibrated ensemble prediction cannot exceed the average performance of its members.
New method learns convolution-like structures from scratch.
problem Learning convolution-like structures from data.
method Minimum description length principle and β-LASSO algorithm. result Learned architectures achieve state-of-the-art accuracies.
We propose sequenced-replacement sampling (SRS) for training deep neural networks. The basic idea is to assign a fixed sequence index to each sample in the dataset. Once a mini-batch is randomly drawn in each training iteration, we refill the original dataset by successively adding samples according to their sequence i…
The study shows removing fully connected output layers improves efficiency without sacrificing performance.
problem Large number of parameters in fully connected layers for high-category datasets.
method Examined architectures replacing fully connected output layers with fixed layers and compared performance.
result Fixed classifiers offer no additional benefit over removing the output layer and its parameters.
Paper proposes CCVR to improve classifier calibration in federated learning with non-IID data.
problem Improving classifier calibration in federated learning with non-IID data.
method Proposes CCVR, a simple algorithm that adjusts the classifier using virtual representations.
result CCVR achieves state-of-the-art performance on popular federated learning benchmarks.
Several recent deep neural networks experiments leverage the generalist-specialist paradigm for classification. However, no formal study compared the performance of different clustering algorithms for class assignment. In this paper we perform such a study, suggest slight modifications to the clustering procedures, and…
TrustNet robustly learns noise patterns from trusted data to improve weakly-supervised classification.
problem Robustness to label noise in weakly-supervised learning.
method TrustNet learns noise patterns from trusted data, then trains a robust classifier using these patterns.
result TrustNet outperforms state-of-the-art methods in robustness to various noise patterns.
Object recognition systems usually require fully complete manually labeled training data to train the classifier. In this paper, we study the problem of object recognition where the training samples are missing during the classifier learning stage, a task also known as zero-shot learning. We propose a novel zero-shot l…
Paper analyzes trade-offs in top-k classification accuracies and proposes a new loss function.
problem CE loss does not always optimize top-k prediction, especially with complex data.
method Introduces a novel top-k transition loss to improve top-k accuracy.
result Our loss function improves top-k accuracy, especially for k > 10.
AANets balance stability and plasticity in CIL.
problem Stability-plasticity dilemma in class-incremental learning.
method Adaptive Aggregation Networks (AANets) with stable and plastic residual blocks.
result AANets improve performance on CIL benchmarks.
Generative models improve adversarial robustness by adding synthetic data.
problem Improving robustness in machine learning models trained on limited data.
method Using synthetic data generated from a large dataset to augment the original training set.
result Generative models can significantly reduce the robust-accuracy gap compared to models trained with additional real data.
A new mutual information optimization method using self-supervised binary contrastive learning.
problem Improving self-supervised contrastive learning for better model performance.
method Proposes a novel loss function for contrastive learning that optimizes mutual information in positive and negative pairs.
result The proposed method outperforms state-of-the-art self-supervised contrastive frameworks on various benchmark datasets.
We consider the problem of designing models to leverage a recently introduced approximate model averaging technique called dropout. We define a simple new model called maxout (so named because its output is the max of a set of inputs, and because it is a natural companion to dropout) designed to both facilitate optimiz…
We present a probabilistic variant of the recently introduced maxout unit. The success of deep neural networks utilizing maxout can partly be attributed to favorable performance under dropout, when compared to rectified linear units. It however also depends on the fact that each maxout unit performs a pooling operation…
Study on top-k classification with new loss functions and algorithms.
problem Improving multi-class classification accuracy and cardinality trade-off.
method Introducing cardinality-aware loss functions and deriving their consistency bounds.
result New cardinality-aware algorithms for top-k classification. Paper introduces new loss functions for multi-class abstention learning.
problem Learning with multi-class classification and the ability to abstain.
method Developed new families of surrogate losses for abstention.
result Proved strong consistency guarantees for new surrogate losses.
Stochastic neural ODEs outperform deterministic ones on image classification tasks.
problem Improving generalization in continuous-time models like neural ODEs.
method Empirical study of stochastically regularized neural ODEs using SDEs.
result Data augmentation negates the benefits of stochastic regularization, making neural ODEs and SDEs nearly equivalent.
Our proposed deeply-supervised nets (DSN) method simultaneously minimizes classification error while making the learning process of hidden layers direct and transparent. We make an attempt to boost the classification performance by studying a new formulation in deep networks. Three aspects in convolutional neural netwo…
Energy-efficient detection of natural errors in deep networks.
problem Deep networks lack error detection capability without additional energy costs.
method Append RACs at hidden layers to detect natural errors with early classification termination.
result Early classification termination reduces energy consumption.
Dual Student separates the teacher from the student in SSL, improving performance.
problem Performance bottleneck caused by coupled teacher in consistency-based SSL methods.
method Introduces Dual Student, replacing the teacher with another student and defining a stabilization constraint.
result Significant improvement in classification performance on SSL benchmarks.
Goldilocks activation functions improve neural network performance.
problem Improving neural network performance and understanding signal transformation.
method Introducing Goldilocks activation functions that locally deform input signals.
result Goldilocks networks outperform or match SELU and RELU on CIFAR-10 and CIFAR-100 datasets.
Subset pretraining speeds up neural network training.
problem Efficiently training neural networks with partial gradients.
method Use small subsets to approximate the loss minimum of the full training set.
result Subset minima can approximate the full training set's loss minimum efficiently.
Max-Cut decision tree improves classification accuracy and reduces computation time.
problem Improving decision tree accuracy and efficiency for complex classification tasks.
method Alternative splitting metric (max cut) and PCA-based feature selection at each node.
result 49% improvement in accuracy with 94% reduction in CPU time on CIFAR-100 data.
In this paper, we introduce Channel-wise recurrent convolutional neural networks (RecNets), a family of novel, compact neural network architectures for computer vision tasks inspired by recurrent neural networks (RNNs). RecNets build upon Channel-wise recurrent convolutional (CRC) layers, a novel type of convolutional …
Most artificial intelligence models have limiting ability to solve new tasks faster, without forgetting previously acquired knowledge. The recently emerging paradigm of continual learning aims to solve this issue, in which the model learns various tasks in a sequential fashion. In this work, a novel approach for contin…
Latest algorithms for automatic neural architecture search perform remarkable but are basically directionless in search space and computational expensive in training of every intermediate architecture. In this paper, we propose a method for efficient architecture search called EENA (Efficient Evolution of Neural Archit…
Proposes a new method to approximate Gaussian inference in classification tasks.
problem Uncertainty quantification in classification tasks using softmax functions.
method Develops a new formalism to approximate Gaussian distributions over logit space and proposes using element-wise normCDF or sigmoid instead of softmax.
result Improves uncertainty quantification compared to softmax Monte Carlo sampling.
Motivated by an important insight from neural science, we propose a new framework for understanding the success of the recently proposed "maxout" networks. The framework is based on encoding information on sparse pathways and recognizing the correct pathway at inference time. Elaborating further on this insight, we pro…
New distance metric for neural architecture search reduces search space complexity.
problem Reducing the complexity of neural architecture search.
method Fisher task distance for measuring task similarity and online neural architecture search.
result Reduced search space complexity for task-specific architectures.
CutMix training technique improves spatial locality in Vision Transformers.
problem Improving spatial locality in Vision Transformers trained from scratch.
method Comparison of Baseline and Modern training protocols on CIFAR-10, CIFAR-100, and Tiny-ImageNet.
result CutMix training component significantly reduces Mean Attention Distance (MAD) in early layers of Vision Transformers.
We proposed the expected energy-based restricted Boltzmann machine (EE-RBM) as a discriminative RBM method for classification. Two characteristics of the EE-RBM are that the output is unbounded and that the target value of correct classification is set to a value much greater than one. In this study, by adopting featur…
SAD-DPSGD improves model performance on imbalanced medical datasets like HAM10000.
problem Data leakage and imbalanced distribution in medical image classification datasets.
method SAD-DPSGD uses a linear decaying mechanism for noise and clipping thresholds to enhance performance.
result SAD-DPSGD outperforms Auto-DPSGD on HAM10000, improving accuracy by 2.15%.
Simplified non-contrastive learning avoids representation collapse.
problem Training failure modes in self-supervised learning.
method Hyperdimensional computing and inductive bias.
result The approach avoids representation collapses.
Autoencoder reduces memory overhead in incremental class learning.
problem Catastrophic forgetting in continual learning.
method Uses autoencoder to extract prototypes and store mean prototypes per class.
result Comparable performance to state-of-the-art with lower memory cost.
CNAS optimizes neural architectures for class-incremental learning.
problem Capacity saturation in static neural architectures for class-incremental learning.
method CNAS uses reinforcement learning and network transformations to adaptively select architectures.
result CNAS outperforms static architectures and is more efficient.
Annotating the right data for training deep neural networks is an important challenge. Active learning using uncertainty estimates from Bayesian Neural Networks (BNNs) could provide an effective solution to this. Despite being theoretically principled, BNNs require approximations to be applied to large-scale problems, …
Adaptive label regularization improves neural network performance.
problem Improving deep neural network performance through label regularization.
method Proposes an adaptive label regularization method that updates label representation online.
result Significant improvement in image recognition and text classification tasks.
We introduce a novel uncertainty estimation for classification tasks for Bayesian convolutional neural networks with variational inference. By normalizing the output of a Softplus function in the final layer, we estimate aleatoric and epistemic uncertainty in a coherent manner. The intractable posterior probability dis…