Related papers: Implicit Semantic Data Augmentation for Deep Netwo…
Deep neural networks with remarkably strong generalization performances are usually over-parameterized. Despite explicit regularization strategies are used for practitioners to avoid over-fitting, the impacts are often small. Some…
Data augmentation (DA) is an essential technique for training state-of-the-art deep learning systems. In this paper, we empirically show data augmentation might introduce noisy augmented examples and consequently hurt the performance on…
The emergence of in-context learning (ICL) enables large pre-trained language models (PLMs) to make predictions for unseen inputs without updating parameters. Despite its potential, ICL's effectiveness heavily relies on the quality,…
Generating surgical reports aimed at surgical scene understanding in robot-assisted surgery can contribute to documenting entry tasks and post-operative analysis. Despite the impressive outcome, the deep learning model degrades the…
In the face of the deep learning model's vulnerability to domain shift, source-free domain adaptation (SFDA) methods have been proposed to adapt models to new, unseen target domains without requiring access to source domain data. Although…
The goal of this thesis is to improve our understanding of the internal mechanisms by which deep artificial neural networks create meaningful representations and are able to generalize. We focus on the challenge of characterizing the…
Due to the difficulty in acquiring massive task-specific occluded images, the classification of occluded images with deep convolutional neural networks (CNNs) remains highly challenging. To alleviate the dependency on large-scale occluded…
Diffusion models has emerged as a powerful framework for tasks like image controllable generation and dense prediction. However, existing models often struggle to capture underlying semantics (e.g., edges, textures, shapes) and effectively…
In this paper, we present an Improved Data Augmentation (IDA) technique focused on Salient Object Detection (SOD). Standard data augmentation techniques proposed in the literature, such as image cropping, rotation, flipping, and resizing,…
Deep learning systems are known to exhibit implicit regularization (alt. implicit bias), favoring simple solutions instead of merely minimizing the loss function. In some cases, we can analytically derive the implicit regularization --…
Designing learning systems which are invariant to certain data transformations is critical in machine learning. Practitioners can typically enforce a desired invariance on the trained model through the choice of a network architecture, e.g.…
For the task of image classification, neural networks primarily rely on visual patterns. In robust networks, we would expect for visually similar classes to be represented similarly. We consider the problem of when semantically similar…
Data augmentation is an effective and universal technique for improving generalization performance of deep neural networks. It could enrich diversity of training samples that is essential in medical image segmentation tasks because 1) the…
Saliency methods have been widely used to highlight important input features in model predictions. Most existing methods use backpropagation on a modified gradient function to generate saliency maps. Thus, noisy gradients can result in…
Transferability estimation identifies the best pre-trained models for downstream tasks without incurring the high computational cost of full fine-tuning. This capability facilitates deployment and advances the pre-training and fine-tuning…
Modern deep learning models generalize remarkably well in-distribution, despite being overparametrized and trained with little to no explicit regularization. Instead, current theory credits implicit regularization imposed by the choice of…
Recent studies on semi-supervised semantic segmentation (SSS) have seen fast progress. Despite their promising performance, current state-of-the-art methods tend to increasingly complex designs at the cost of introducing more network…
Most previous methods for text data augmentation are limited to simple tasks and weak baselines. We explore data augmentation on hard tasks (i.e., few-shot natural language understanding) and strong baselines (i.e., pretrained models with…
Knowledge distillation (KD) is an efficient framework for compressing large-scale pre-trained language models. Recent years have seen a surge of research aiming to improve KD by leveraging Contrastive Learning, Intermediate Layer…
Efficient deployment of deep neural networks on resource-constrained devices demands advanced compression techniques that preserve accuracy and interoperability. This paper proposes a machine learning framework that augments Knowledge…