English
Related papers

Related papers: Improving the Straight-Through Estimator with Zero…

200 papers

Fractional gradient descent has been studied extensively, with a focus on its ability to extend traditional gradient descent methods by incorporating fractional-order derivatives. This approach allows for more flexibility in navigating…

Machine Learning · Computer Science 2024-11-25 Teodor Alexandru Szente , James Harrison , Mihai Zanfir , Cristian Sminchisescu

Stochastic First-Order (SFO) methods have been a cornerstone in addressing a broad spectrum of modern machine learning (ML) challenges. However, their efficacy is increasingly questioned, especially in large-scale applications where…

Machine Learning · Computer Science 2024-08-01 Di Zhang , Suvrajeet Sen

Gradient descent and its variants are widely used in machine learning. However, oracle access of gradient may not be available in many applications, limiting the direct use of gradient descent. This paper proposes a method of estimating…

Optimization and Control · Mathematics 2019-10-07 Qinbo Bai , Mridul Agarwal , Vaneet Aggarwal

We introduce LOREN, a curvature-aware zeroth-order (ZO) optimization method for fine-tuning large language models (LLMs). Existing ZO methods, which estimate gradients via finite differences using random perturbations, often suffer from…

Machine Learning · Computer Science 2025-11-12 Hyunseok Seung , Jaewoo Lee , Hyunsuk Ko

Fine-tuning pretrained self-supervised language models is widely adopted for transfer learning to downstream tasks. Fine-tuning can be achieved by freezing gradients of the pretrained network and only updating gradients of a newly added…

Computation and Language · Computer Science 2023-07-20 James O' Neill , Sourav Dutta

Pre-training has been investigated to improve the efficiency and performance of training neural operators in data-scarce settings. However, it is largely in its infancy due to the inherent complexity and diversity, such as long…

Machine Learning · Computer Science 2024-05-08 Zhongkai Hao , Chang Su , Songming Liu , Julius Berner , Chengyang Ying , Hang Su , Anima Anandkumar , Jian Song , Jun Zhu

Forward Gradients - the idea of using directional derivatives in forward differentiation mode - have recently been shown to be utilizable for neural network training while avoiding problems generally associated with backpropagation gradient…

Machine Learning · Computer Science 2023-06-13 Louis Fournier , Stéphane Rivaud , Eugene Belilovsky , Michael Eickenberg , Edouard Oyallon

Fine-tuning Large Language Models (LLMs) with first-order methods like back-propagation is computationally intensive. Zeroth-Order (ZO) optimisation uses function evaluations instead of gradients, reducing memory usage, but suffers from…

Computation and Language · Computer Science 2025-07-24 Alessio Galatolo , Zhenbang Dai , Katie Winkle , Meriem Beloucif

Online minimization of an unknown convex function over the interval $[0,1]$ is considered under first-order stochastic bandit feedback, which returns a random realization of the gradient of the function at each query point. Without knowing…

Machine Learning · Statistics 2020-02-21 Sattar Vakili , Sudeep Salgia , Qing Zhao

Stochastic gradient descent method and its variants constitute the core optimization algorithms that achieve good convergence rates for solving machine learning problems. These rates are obtained especially when these algorithms are…

Machine Learning · Computer Science 2024-03-14 S. Ilker Birbil , Ozgur Martin , Gonenc Onay , Figen Oztoprak

Differentiable architecture search (DARTS) has been a popular one-shot paradigm for NAS due to its high efficiency. It introduces trainable architecture parameters to represent the importance of candidate operations and proposes…

Computer Vision and Pattern Recognition · Computer Science 2022-02-01 Xiaoxing Wang , Wenxuan Guo , Junchi Yan , Jianlin Su , Xiaokang Yang

Fully quantized training (FQT), which uses low-bitwidth hardware by quantizing the activations, weights, and gradients of a neural network model, is a promising approach to accelerate the training of deep neural networks. One major…

Machine Learning · Computer Science 2020-10-28 Jianfei Chen , Yu Gai , Zhewei Yao , Michael W. Mahoney , Joseph E. Gonzalez

Quantization is an effective technique to reduce memory footprint, inference latency, and power consumption of deep learning models. However, existing quantization methods suffer from accuracy degradation compared to full-precision (FP)…

Machine Learning · Computer Science 2022-10-14 Zheng Wang , Juncheng B Li , Shuhui Qu , Florian Metze , Emma Strubell

This paper provides a new way of developing the fast iterative shrinkage/thresholding algorithm (FISTA) that is widely used for minimizing composite convex functions with a nonsmooth term such as the $\ell_1$ regularizer. In particular,…

Optimization and Control · Mathematics 2019-06-14 Donghwan Kim , Jeffrey A. Fessler

In finetuning a large pretrained model to downstream tasks, parameter-efficient fine-tuning (PEFT) methods can effectively finetune pretrained models with few trainable parameters, but suffer from high GPU memory consumption and slow…

Computer Vision and Pattern Recognition · Computer Science 2024-02-07 Ningyuan Tang , Minghao Fu , Ke Zhu , Jianxin Wu

The analysis of gradient descent-type methods typically relies on the Lipschitz continuity of the objective gradient. This generally requires an expensive hyperparameter tuning process to appropriately calibrate a stepsize for a given…

Optimization and Control · Mathematics 2023-11-16 Albert S. Berahas , Lindon Roberts , Fred Roosta

Quantization reduces computation costs of neural networks but suffers from performance degeneration. Is this accuracy drop due to the reduced capacity, or inefficient training during the quantization procedure? After looking into the…

Computer Vision and Pattern Recognition · Computer Science 2019-12-24 Qing Jin , Linjie Yang , Zhenyu Liao

Distributed optimization is fundamental to modern machine learning applications like federated learning, but existing methods often struggle with ill-conditioned problems and face stability-versus-speed tradeoffs. We introduce fractional…

Machine Learning · Computer Science 2024-12-04 Andrei Lixandru , Marcel van Gerven , Sergio Pequito

Decentralized strategies are of interest for learning from large-scale data over networks. This paper studies learning over a network of geographically distributed nodes/agents subject to quantization. Each node possesses a private local…

Systems and Control · Electrical Eng. & Systems 2025-01-13 Mohammadreza Doostmohammadian , Muhammad I. Qureshi , Mohammad Hossein Khalesi , Hamid R. Rabiee , Usman A. Khan

The concept of learning to optimize involves utilizing a trainable optimization strategy rather than relying on manually defined full gradient estimations such as ADAM. We present a framework that jointly trains the full gradient estimator…

Machine Learning · Computer Science 2026-01-30 Ruiqi Wang , Diego Klabjan