Skip to search formSkip to main contentSkip to account menu

Densely Connected Convolutional Networks

The Dense Convolutional Network (DenseNet), which connects each layer to every other layer in a feed-forward fashion, and has several compelling advantages: they alleviate the vanishing-gradient problem, strengthen feature propagation, encourage feature reuse, and substantially reduce the number of parameters.
View on IEEE (opens in a new tab)

Learning Efficient Convolutional Networks through Network Slimming

The approach is called network slimming, which takes wide and large networks as input models, but during training insignificant channels are automatically identified and pruned afterwards, yielding thin and compact models with comparable accuracy.
View on IEEE (opens in a new tab)

Deep Networks with Stochastic Depth

Stochastic depth is proposed, a training procedure that enables the seemingly contradictory setup to train short networks and use deep networks at test time and reduces training time substantially and improves the test error significantly on almost all data sets that were used for evaluation.
View PDF on arXiv (opens in a new tab)

A ConvNet for the 2020s

This work gradually “modernize” a standard ResNet toward the design of a vision Transformer, and discovers several key components that contribute to the performance difference along the way, leading to a family of pure ConvNet models dubbed ConvNeXt.
View on IEEE (opens in a new tab)

Rethinking the Value of Network Pruning

It is found that with optimal learning rate, the "winning ticket" initialization as used in Frankle & Carbin (2019) does not bring improvement over random initialization, and the need for more careful baseline evaluations in future research on structured pruning methods is suggested.
View PDF on arXiv (opens in a new tab)

Few-Shot Object Detection via Feature Reweighting

This work develops a few-shot object detector that can learn to detect novel objects from only a few annotated examples, using a meta feature learner and a reweighting module within a one-stage detection architecture.
View on IEEE (opens in a new tab)

Snapshot Ensembles: Train 1, get M for free

This paper proposes a method to obtain the seemingly contradictory goal of ensembling multiple neural networks at no additional training cost by training a single neural network, converging to several local minima along its optimization path and saving the model parameters.
View PDF on arXiv (opens in a new tab)

DSOD: Learning Deeply Supervised Object Detectors from Scratch

Deeply Supervised Object Detector (DSOD), a framework that can learn object detectors from scratch following the single-shot detection (SSD) framework, and one of the key findings is that deep supervision, enabled by dense layer-wise connections, plays a critical role in learning a good detector.
View on IEEE (opens in a new tab)

Test-Time Training with Self-Supervision for Generalization under Distribution Shifts

This work turns a single unlabeled test sample into a self-supervised learning problem, on which the model parameters are updated before making a prediction, which leads to improvements on diverse image classification benchmarks aimed at evaluating robustness to distribution shifts.
View Paper (opens in a new tab)

A New Meta-Baseline for Few-Shot Learning

This work presents a Meta-Baseline method, by pre-training a classifier on all base classes and meta-learning on a nearest-centroid based few-shot classification algorithm, which outperforms recent state-of-the-art methods by a large margin.
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)