English
Related papers

Related papers: Analyzing Redundancy in Pretrained Transformer Mod…

200 papers

Recurrent Neural Networks (RNNs) are frequently used to model aspects of brain function and structure. In this work, we trained small fully-connected RNNs to perform temporal and flow control tasks with time-varying stimuli. Our results…

Neurons and Cognition · Quantitative Biology 2023-06-29 Cecilia Jarne , Rodrigo Laje

The increasing scale of modern neural networks, exemplified by architectures from IBM (530 billion neurons) and Google (500 billion parameters), presents significant challenges in terms of computational cost and infrastructure requirements.…

Machine Learning · Computer Science 2025-06-03 Paritosh Ranjan , Surajit Majumder , Prodip Roy

We rigorously evaluate three state-of-the-art techniques for inducing sparsity in deep neural networks on two large-scale learning tasks: Transformer trained on WMT 2014 English-to-German, and ResNet-50 trained on ImageNet. Across thousands…

Machine Learning · Computer Science 2019-02-27 Trevor Gale , Erich Elsen , Sara Hooker

The computation and storage requirements for Deep Neural Networks (DNNs) are usually high. This issue limits their deployability on ubiquitous computing devices such as smart phones, wearables and autonomous drones. In this paper, we…

Machine Learning · Computer Science 2017-02-28 Hande Alemdar , Vincent Leroy , Adrien Prost-Boucle , Frédéric Pétrot

Adapting pre-trained neural models to downstream tasks has become the standard practice for obtaining high-quality models. In this work, we propose a novel model adaptation paradigm, adapting by pruning, which prunes neural connections in…

Machine Learning · Computer Science 2021-05-10 Yang Gao , Nicolo Colombo , Wei Wang

In this report, we combine the idea of Wide ResNets and transfer learning to optimize the architecture of deep neural networks. The first improvement of the architecture is the use of all layers as information source for the last layer.…

Machine Learning · Computer Science 2022-06-22 Wolfgang Fuhl

Convolutional neural networks (CNN) are generally designed with a heuristic initialization of network architecture and trained for a certain task. This often leads to overparametrization after learning and induces redundancy in the…

Machine Learning · Computer Science 2019-06-11 Rachana Sathish , Debdoot Sheet

Recurrent neural networks (RNNs) are widely used as a memory model for sequence-related problems. Many variants of RNN have been proposed to solve the gradient problems of training RNNs and process long sequences. Although some classical…

Neural and Evolutionary Computing · Computer Science 2020-05-29 Chenpeng Zhang , Shuai Li , Mao Ye , Ce Zhu , Xue Li

Transfer learning has emerged as a powerful methodology for adapting pre-trained deep neural networks on image recognition tasks to new domains. This process consists of taking a neural network pre-trained on a large feature-rich source…

Machine Learning · Computer Science 2021-04-27 Francisco Utrera , Evan Kravitz , N. Benjamin Erichson , Rajiv Khanna , Michael W. Mahoney

Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance in many computer vision tasks over the years. However, this comes at the cost of heavy computation and memory intensive network designs, suggesting potential…

Computer Vision and Pattern Recognition · Computer Science 2020-08-11 Kumara Kahatapitiya , Ranga Rodrigo

Modern deep convolutional networks (CNNs) are often criticized for not generalizing under distributional shifts. However, several recent breakthroughs in transfer learning suggest that these networks can cope with severe distribution shifts…

In this work, we study how well the learned weights of a neural network utilize the space available to them. This notion is related to capacity, but additionally incorporates the interaction of the network architecture with the dataset.…

Computer Vision and Pattern Recognition · Computer Science 2024-07-09 Isha Garg , Christian Koguchi , Eshan Verma , Daniel Ulbricht

Deep neural networks (DNNs) are powerful learning machines that have enabled breakthroughs in several domains. In this work, we introduce a new retrospective loss to improve the training of deep neural network models by utilizing the prior…

Computer Vision and Pattern Recognition · Computer Science 2020-06-25 Surgan Jandial , Ayush Chopra , Mausoom Sarkar , Piyush Gupta , Balaji Krishnamurthy , Vineeth Balasubramanian

Transfer learning is a popular practice in deep neural networks, but fine-tuning of large number of parameters is a hard task due to the complex wiring of neurons between splitting layers and imbalance distributions of data in pretrained…

Machine Learning · Computer Science 2017-10-23 Arash Shahriari

Natural Language Processing (NLP) has witnessed a transformative leap with the advent of transformer-based architectures, which have significantly enhanced the ability of machines to understand and generate human-like text. This paper…

Computation and Language · Computer Science 2025-03-27 Tianhao Wu , Yu Wang , Ngoc Quach

Redundancy is a fundamental characteristic of many biological processes such as those in the genetic, visual, muscular and nervous system; yet its function has not been fully understood. The conventional interpretation of redundancy is that…

Information Theory · Computer Science 2019-02-04 Anh Tuan Nguyen , Jian Xu , Diu Khue Luu , Qi Zhao , Zhi Yang

The practical success of much of NLP depends on the availability of training data. However, in real-world scenarios, training data is often scarce, not least because many application domains are restricted and specific. In this work, we…

Computation and Language · Computer Science 2022-04-01 Marina Sedinkina , Martin Schmitt , Hinrich Schütze

Fine-tuning a pre-trained model, such as Bidirectional Encoder Representations from Transformers (BERT), has been proven to be an effective method for solving many natural language processing (NLP) tasks. However, due to the large number of…

Computation and Language · Computer Science 2024-04-01 Taha ValizadehAslani , Hualou Liang

Large pre-trained language models help to achieve state of the art on a variety of natural language processing (NLP) tasks, nevertheless, they still suffer from forgetting when incrementally learning a sequence of tasks. To alleviate this…

Computation and Language · Computer Science 2023-03-03 Mingxu Tao , Yansong Feng , Dongyan Zhao

Recent advancements in the NLP field showed that transfer learning helps with achieving state-of-the-art results for new tasks by tuning pre-trained models instead of starting from scratch. Transformers have made a significant improvement…

Computation and Language · Computer Science 2020-09-14 Aysu Ezen-Can