The study finds a theoretical bound for pre-training iterations needed for pruning to yield good subnetwork performance.
problem Discovering efficient subnetworks within pre-trained dense networks.
method Mathematical analysis of a two-layer, fully-connected network, validating with a multi-layer perceptron trained on MNIST.
result A logarithmically dependent threshold on dataset size for successful pruning.
Paper shows pre-training and transfer learning reduce sample complexity for neural networks.
problem Training high-dimensional supervised learning with limited labeled data.
method Study of single-layer neural networks via online stochastic gradient descent, considering concept shift.
result Pre-training and transfer learning reduce sample complexity by polynomial factors under general assumptions.
GCC pre-trains graph neural networks to transfer across diverse datasets.
problem Non-transferable graph models trained for specific datasets.
method Graph Contrastive Coding (GCC) framework using contrastive learning.
result GCC pre-trained models achieve competitive performance across multiple datasets.
This work proves composite neural networks outperform their components under certain conditions.
problem Understanding the performance of composite neural networks.
method Theoretical investigation of a composite neural network combining pre-trained models.
result A composite neural network, with high probability, performs better than any of its pre-trained components.
We present a novel approach to leverage large unlabeled datasets by pre-training state-of-the-art deep neural networks on randomly-labeled datasets. Specifically, we train the neural networks to memorize arbitrary labels for all the samples in a dataset and use these pre-trained networks as a starting point for regular…
Pre-training is crucial for learning deep neural networks. Most of existing pre-training methods train simple models (e.g., restricted Boltzmann machines) and then stack them layer by layer to form the deep structure. This layer-wise pre-training has found strong theoretical foundation and broad empirical support. Howe…
Behavior Transfer improves reinforcement learning by leveraging pre-trained policies.
problem Efficient transfer of knowledge in reinforcement learning.
method Behavior Transfer (BT) that uses pre-trained policies for exploration.
result BT combined with pre-training leads to better solutions than without pre-training.
Survey on self-supervised pre-training for neural networks using unlabeled data.
problem Improving model performance using unlabeled data.
method Pre-training on unlabeled data followed by task-specific adaptation.
result Enhanced model performance through self-supervised pre-training.
SimCLR pre-training improves CNN performance with fewer labels.
problem Learning with fewer labeled data.
method SimCLR contrastive learning method combined with supervised fine-tuning.
result SimCLR pre-training with supervised fine-tuning achieves almost optimal test loss with fewer labeled data.
Adaptive kernel density estimation improves accuracy in high dimensions.
problem Challenges in high-dimensional density estimation with traditional methods.
method Pre-training a neural network to recommend location-adaptive kernels.
result Effective density estimation in high dimensions with improved accuracy.
GPT-GNN pre-trains GNNs on unlabeled graphs to improve downstream performance.
problem Training GNNs requires labeled data, which is expensive.
method Generative pre-training of GNNs on unlabeled data with self-supervision.
result GPT-GNN significantly outperforms state-of-the-art GNNs without pre-training.
Many applications of machine learning require a model to make accurate pre-dictions on test examples that are distributionally different from training ones, while task-specific labels are scarce during training. An effective approach to this challenge is to pre-train a model on related tasks where data is abundant, and…
Graph neural networks (GNNs) are shown to be successful in modeling applications with graph structures. However, training an accurate GNN model requires a large collection of labeled data and expressive features, which might be inaccessible for some applications. To tackle this problem, we propose a pre-training framew…
We propose a novel learning method for multilayered neural networks which uses feedforward supervisory signal and associates classification of a new input with that of pre-trained input. The proposed method effectively uses rich input information in the earlier layer for robust leaning and revising internal representat…
Composite neural network improves PM2.5 prediction.
problem Improving PM2.5 prediction accuracy.
method Composite neural network framework with pre-trained models.
result Composite neural network outperforms individual models and new components added.
This paper ranks pre-trained DNNs using a novel SI measure.
problem Optimizing pre-trained DNN selection for transfer learning.
method Automated ranking via Separation Index (SI) on target datasets.
result Ranked pre-trained DNNs improve classification performance.
We consider unsupervised domain adaptation: given labelled examples from a source domain and unlabelled examples from a related target domain, the goal is to infer the labels of target examples. Under the assumption that features from pre-trained deep neural networks are transferable across related domains, domain adap…
Study compares deep feature methods for anomaly detection in limited data scenarios.
problem Handling limited data in industrial inspection applications.
method Three approaches (KNN, Mahalanobis, PaDiM) using pre-trained deep features with data augmentation.
result Data augmentation significantly improves performance in small data regimes.
New method improves ensemble quality by exploring the pre-train basin more effectively.
problem Limited diversity in ensembles trained from a single pre-trained checkpoint.
method Proposed StarSSE modification of Snapshot Ensembles for transfer learning.
result Stronger ensembles and uniform model soups achieved.
SAINT improves neural networks for tabular data with row attention and contrastive pre-training.
problem Tabular data challenges in machine learning applications.
method SAINT combines row and column attention with contrastive self-supervised pre-training.
result SAINT outperforms previous deep learning methods and even gradient boosting methods on benchmark tasks.
Post-hoc uncertainty quantification improves on pre-trained neural networks without underfitting.
problem Uncertainty quantification in neural networks is underfitting or computationally demanding.
method Gaussian Process Activation function (GAPA) for neuron-level uncertainty, with two methods: GAPA-Free and GAPA-Variational.
result GAPA-Variational outperforms Laplace approximation on most datasets in uncertainty quantification metrics.
Theoretical study shows adversarial training improves robustness in deep learning models.
problem Ensuring robustness in pre-trained deep learning models.
method Theoretical analysis of adversarial training and feature purification in two-layer neural networks.
result Adversarial training leads to feature purification, making models more robust to attacks.
TCGPN improves stock forecasting by capturing temporal correlation patterns.
problem Stock forecasting with minimal periodicity and large node numbers.
method TCGPN uses Temporal-Correlation fusion encoder and pre-training methods to handle large datasets.
result TCGPN achieves state-of-the-art results on real stock market data.
Unsupervised pre-training improves model generalization, but lacks theoretical understanding.
problem Lack of theoretical understanding of unsupervised pre-training's impact on model generalization.
method Introduces a novel theoretical framework to analyze and enhance generalization.
result Enhances understanding of unsupervised pre-training and fine-tuning, proposing a new regularization method.
Deep Neural Networks (DNNs) have become increasingly popular in computer vision, natural language processing, and other areas. However, training and fine-tuning a deep learning model is computationally intensive and time-consuming. We propose a new method to improve the performance of nearly every model including pre-t…
Ensemble pre-trained models for low data transfer learning.
problem Training good models from scratch in low data regime.
method Fine-tune nearest-neighbour ranked pre-trained models to create ensembles.
result Achieves state-of-the-art performance with lower inference budget.
Improved LLM pre-training performance through better weight and variance control.
problem Improper weight and variance control in LLM pre-training affects downstream task performance.
method Introduced Layer Index Rescaling (LIR) and Target Variance Rescaling (TVR) techniques.
result Substantial improvements in downstream task performance (up to 4.6%) and reduced extreme activation values.
Personalizes pre-trained models for nonparametric regression with limited data.
problem Improving data efficiency in nonparametric regression with few samples.
method Develops a theoretical framework and algorithms for few-shot personalization of black-box models.
result Achieves minimax optimal rate for personalization in nonparametric regression.
PLUS pre-trains protein sequences with structural info, improving performance.
problem Lack of labeled protein sequences for training models.
method PLUS combines masked language modeling with same-family prediction for pre-training.
result PLUS-RNN outperforms other models in protein biology tasks.
The study uses pre-trained neural networks to adjust for confounding in non-tabular data.
problem Neglecting non-tabular data sources can lead to biased ATE estimates.
method Leverages latent features from pre-trained neural networks to adjust for confounding.
result Neural networks can achieve fast convergence rates for ATE estimation with latent features.
PFNs pre-train models on simulated data to predict class probabilities.
problem Training machine learning models on large datasets.
method Pre-train a fixed model on small simulated datasets and use it to infer class probabilities in-context.
result PFNs achieve state-of-the-art performance and improve with larger inference data.
Unbalanced GANs stabilize GAN training by pre-training the generator with VAE.
problem Stable training of GANs to avoid mode collapses and improve image quality.
method Pre-train GAN generator with VAE, balance generator and discriminator training, prevent discriminator's early convergence.
result Unbalanced GANs reduce mode collapses and outperform ordinary GANs in stability, convergence, and image quality.
Deep convolutional neural networks have achieved great success in various applications. However, training an effective DNN model for a specific task is rather challenging because it requires a prior knowledge or experience to design the network architecture, repeated trial-and-error process to tune the parameters, and …
Deep neural networks have shown promising results for various clinical prediction tasks. However, training deep networks such as those based on Recurrent Neural Networks (RNNs) requires large labeled data, significant hyper-parameter tuning effort and expertise, and high computational resources. In this work, we invest…
Lottery tickets find good initializations for IMP with sparse training.
problem Finding good initializations for iterative magnitude pruning (IMP) in sparse networks.
method Empirical study of IMP performance with varying pre-training data and iterations.
result Training on a small fraction of data suffices to obtain good initializations for IMP.
This paper aims to analyze knowledge consistency between pre-trained deep neural networks. We propose a generic definition for knowledge consistency between neural networks at different fuzziness levels. A task-agnostic method is designed to disentangle feature components, which represent the consistent knowledge, from…
Fine-tunes GNNs by preserving generative patterns to improve transferability.
problem Vanilla fine-tuning fails due to structural divergence between pre-training and downstream graphs.
method G-Tuning, which reconstructs the generative patterns of the downstream graph using graphon bases.
result G-Tuning achieves an average improvement of 0.5% and 2.6% on in-domain and out-of-domain transfer learning experiments.
Enhances deep neural networks with fixed-mean Gaussian processes for uncertainty estimation.
problem Post-hoc uncertainty estimation of pre-trained deep neural networks.
method Fixed-mean Gaussian processes with variational inference for efficient stochastic optimization.
result FMGP improves uncertainty estimation and computational efficiency compared to state-of-the-art methods.
Improves deep transfer learning by preventing performance degradation.
problem Deep transfer learning can degrade performance when using inappropriate pre-trained weights.
method Proposes a novel strategy to compute new descent directions that preserve regularization effects.
result DTNH strategy improves performance of deep transfer learning tasks by 0.1%--7%.
Study improves breast lesion segmentation with limited in vivo data using simulated and natural images.
problem Challenges in automatic breast lesion segmentation due to limited annotated data.
method Pre-training a segmentation network on simulated and natural images, followed by fine-tuning with limited in vivo data.
result Fine-tuning improves dice score by 21% with as little as 19 in vivo images.
New findings show BERT subnetworks can train independently and transfer to various tasks.
problem Finding smaller subnetworks that can train independently and transfer to other tasks.
method Examined pre-trained BERT models for subnetworks that can train independently and transfer to various downstream tasks.
result Found subnetworks at 40% to 90% sparsity that can train independently and transfer to various tasks.
Deep neural networks have achieved impressive performance in many applications but their large number of parameters lead to significant computational and storage overheads. Several recent works attempt to mitigate these overheads by designing compact networks using pruning of connections. However, we observe that most …
LEMON uses pre-trained models to scale neural networks efficiently.
problem Efficiency in scaling deep neural networks, especially Transformers, which are resource-intensive to train from scratch.
method LEMON initializes scaled models using pre-trained weights and optimizes learning rates.
result Significant reduction in training time and computational costs for Vision Transformers and BERT.
In this contribution, we present a novel approach for segmenting laser radar (lidar) imagery into geometric time-height cloud locations with a fully convolutional network (FCN). We describe a semi-supervised learning method to train the FCN by: pre-training the classification layers of the FCN with image-level annotati…
ABNN converts pre-trained DNNs into BNNs for reliable uncertainty quantification.
problem Uncertainty quantification in deep neural networks (DNNs) is challenging and critical for real-world applications.
method Adaptable Bayesian Neural Network (ABNN) that transforms pre-trained DNNs into BNNs with minimal overhead.
result ABNN achieves state-of-the-art performance in image classification and semantic segmentation tasks.
Transfer learning boosts chemically accurate neural network potentials for organic molecules.
problem Developing accurate interatomic potentials from ab-initio data.
method Discriminative fine-tuning of pre-trained neural networks.
result Fine-tuning with energy labels alone can achieve accurate atomic forces.
Deep neural networks have shown promising results for various clinical prediction tasks such as diagnosis, mortality prediction, predicting duration of stay in hospital, etc. However, training deep networks -- such as those based on Recurrent Neural Networks (RNNs) -- requires large labeled data, high computational res…
Capturing sentence semantics plays a vital role in a range of text mining applications. Despite continuous efforts on the development of related datasets and models in the general domain, both datasets and models are limited in biomedical and clinical domains. The BioCreative/OHNLP organizers have made the first attemp…