English
Related papers

Related papers: On orthogonality and learning recurrent networks w…

200 papers

The increasing size of neural networks has led to a growing demand for methods of efficient fine-tuning. Recently, an orthogonal fine-tuning paradigm was introduced that uses orthogonal matrices for adapting the weights of a pretrained…

Machine Learning · Computer Science 2024-06-17 Mikhail Gorbunov , Nikolay Yudin , Vera Soboleva , Aibek Alanov , Alexey Naumov , Maxim Rakhuba

Recurrent neural networks (RNNs) are wide-spread machine learning tools for modeling sequential and time series data. They are notoriously hard to train because their loss gradients backpropagated in time tend to saturate or diverge during…

Machine Learning · Computer Science 2022-10-10 Jonas M. Mikhaeil , Zahra Monfared , Daniel Durstewitz

Recurrent neural networks (RNNs) notoriously struggle to learn long-term memories, primarily due to vanishing and exploding gradients. The recent success of state-space models (SSMs), a subclass of RNNs, to overcome such difficulties…

Machine Learning · Computer Science 2024-11-06 Nicolas Zucchet , Antonio Orvieto

We explain that the difficulties of training deep neural networks come from a syndrome of three consistency issues. This paper describes our efforts in their analysis and treatment. The first issue is the training speed inconsistency in…

Machine Learning · Computer Science 2017-08-03 Chengxi Ye , Yezhou Yang , Cornelia Fermuller , Yiannis Aloimonos

When an online learning algorithm is used to estimate the unknown parameters of a model, the signals interacting with the parameter estimates should not decay too quickly for the optimal values to be discovered correctly. This requirement…

Machine Learning · Computer Science 2019-11-05 Kamil Nar , S. Shankar Sastry

Unitary neural networks are promising alternatives for solving the exploding and vanishing activation/gradient problem without the need for explicit normalization that reduces the inference speed. However, they often require longer training…

Machine Learning · Computer Science 2021-02-22 Hao-Yuan Chang

Vanishing and exploding gradients are two of the main obstacles in training deep neural networks, especially in capturing long range dependencies in recurrent neural networks~(RNNs). In this paper, we present an efficient parametrization of…

Machine Learning · Computer Science 2018-03-28 Jiong Zhang , Qi Lei , Inderjit S. Dhillon

In this paper, acceleration of gradient methods for convex optimization problems with weak levels of convexity and smoothness is considered. Starting from the universal fast gradient method which was designed to be an optimal method for…

Optimization and Control · Mathematics 2022-06-10 Jongho Park

Adversarial robustness is essential for security and reliability of machine learning systems. However, adversarial robustness enhanced by defense algorithms is easily erased as the neural network's weights update to learn new tasks. To…

Machine Learning · Computer Science 2024-08-14 Xiaolei Ru , Xiaowei Cao , Zijia Liu , Jack Murdoch Moore , Xin-Ya Zhang , Xia Zhu , Wenjia Wei , Gang Yan

Continual learning, the ability of a model to adapt to an ongoing sequence of tasks without forgetting earlier ones, is a central goal of artificial intelligence. To better understand its underlying mechanisms, we study the limitations of…

Machine Learning · Statistics 2026-04-21 Hossein Taheri , Avishek Ghosh , Arya Mazumdar

Neural networks have been successfully used for classification tasks in a rapidly growing number of practical applications. Despite their popularity and widespread use, there are still many aspects of training and classification that are…

Machine Learning · Computer Science 2016-05-03 Ewout van den Berg

Recently, orthogonal recurrent neural networks (RNNs) have emerged as state-of-the-art models for learning long-term dependencies. This class of models mitigates the exploding and vanishing gradients problem by design. In this work, we…

Machine Learning · Computer Science 2021-02-22 Omri Azencot , N. Benjamin Erichson , Mirela Ben-Chen , Michael W. Mahoney

Recurrent neural networks (RNNs) are notoriously difficult to train. When the eigenvalues of the hidden to hidden weight matrix deviate from absolute value 1, optimization becomes difficult due to the well studied issue of vanishing and…

Machine Learning · Computer Science 2016-10-13 Martin Arjovsky , Amar Shah , Yoshua Bengio

Recursive neural networks (RNN) and their recently proposed extension recursive long short term memory networks (RLSTM) are models that compute representations for sentences, by recursively combining word embeddings according to an…

Artificial Intelligence · Computer Science 2016-03-02 Phong Le , Willem Zuidema

Orthogonal matrix has shown advantages in training Recurrent Neural Networks (RNNs), but such matrix is limited to be square for the hidden-to-hidden transformation in RNNs. In this paper, we generalize such square orthogonal matrix to…

Machine Learning · Computer Science 2017-11-22 Lei Huang , Xianglong Liu , Bo Lang , Adams Wei Yu , Yongliang Wang , Bo Li

Tree-based models are widely recognized for their interpretability and have proven effective in various application domains, particularly in high-stakes domains. However, learning decision trees (DTs) poses a significant challenge due to…

Machine Learning · Computer Science 2026-03-13 Sascha Marton

Training a neural network using backpropagation algorithm requires passing error gradients sequentially through the network. The backward locking prevents us from updating network layers in parallel and fully leveraging the computing…

Machine Learning · Computer Science 2019-05-30 Zhouyuan Huo , Bin Gu , Heng Huang

Optimization problems with rank constraints arise in many applications, including matrix regression, structured PCA, matrix completion and matrix decomposition problems. An attractive heuristic for solving such problems is to factorize the…

Statistics Theory · Mathematics 2015-09-11 Yudong Chen , Martin J. Wainwright

In recent years, Orthogonal Recurrent Neural Networks (ORNNs) have gained popularity due to their ability to manage tasks involving long-term dependencies, such as the copy-task, and their linear complexity. However, existing ORNNs utilize…

Neural and Evolutionary Computing · Computer Science 2024-06-11 Armand Foucault , Franck Mamalet , François Malgouyres

The optimisation of neural networks can be sped up by orthogonalising the gradients before the optimisation step, ensuring the diversification of the learned representations. We orthogonalise the gradients of the layer's components/filters…

Machine Learning · Computer Science 2022-02-16 Mark Tuddenham , Adam Prügel-Bennett , Jonathan Hare