☆ 4.6 Article

Developing Novel Robust Loss Functions-Based Classification Layers for DLLSTM Neural Networks

IEEE ACCESS (2023)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Computer Science, Artificial Intelligence

ILF-LSTM: enhanced loss function in LSTM to predict the sea surface temperature

Bhimavarapu Usharani

Summary: This research presents improved sea surface temperature prediction using a long short-term memory neural network with an improved loss function. By considering long-term dependencies and extracting features from spatial data, this approach outperforms standard prediction models and other existing techniques.

SOFT COMPUTING (2023)

添加到收藏夹

Review Chemistry, Multidisciplinary

An Overview of Machine Learning, Deep Learning, and Reinforcement Learning-Based Techniques in Quantitative Finance: Recent Progress and Challenges

Santosh Kumar Sahu et al.

Summary: Forecasting stock market behavior is a challenging topic that has attracted the interest of economists and computer scientists. Researchers have explored linear models, as well as machine learning, deep learning, reinforcement learning, and deep reinforcement learning, in order to create accurate predictive models. Machine learning algorithms can identify financial market patterns, and investors are using deep learning models to analyze stock and forex markets with the help of artificial intelligence. Deep reinforcement learning algorithms have also gained popularity in algorithmic trading. Our objective is to provide a comprehensive overview of the utility of ML, DL, RL, and DRL in quantitative finance and the stock market, and suggest potential future research paths based on the existing findings.

APPLIED SCIENCES-BASEL (2023)

添加到收藏夹

Article Mathematics

Machine-Learning Methods on Noisy and Sparse Data

Konstantinos Poulinakis et al.

Summary: This study compares machine-learning methods and cubic splines on their ability to handle sparse and noisy training data. The results show that cubic splines provide more precise interpolation than deep neural networks and multivariate adaptive regression splines with very sparse data. However, machine-learning models show robustness to noise and can outperform splines after reaching a threshold of training data. The study aims to provide a general framework for interpolating one-dimensional signals, often obtained from complex scientific simulations or laboratory experiments.

MATHEMATICS (2023)

添加到收藏夹

Article Chemistry, Multidisciplinary

Long Short-Term Memory Recurrent Neural Network Approach for Approximating Roots (Eigen Values) of Transcendental Equation of Cantilever Beam

Madiha Bukhsh et al.

Summary: In this study, a long short-term memory-recurrent neural network (LSTM-RNN) approach is used to estimate the natural frequencies and roots (Eigenvalues) of a cantilever steel beam for transverse vibration with clamped free (CF) boundary conditions. The finite element method (FEM) package ANSYS is employed for dynamic analysis and the Euler-Bernoulli beam theory is used to generate sample datasets. A deep neural network (DNN)-based LSTM-RNN technique is implemented to approximate the roots of the transcendental equation. The network validity is demonstrated by comparing the numerical and simulated results using the LSTM-RNN methodology.

APPLIED SCIENCES-BASEL (2023)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Performance evaluation of frequency division duplex (FDD) massive multiple input multiple output (MIMO) under different correlation models

Alaa M. Abdul-Hadi et al.

Summary: This article focuses on the technical challenge of downlink channel state information (CSI) estimation in a frequency-division duplex (FDD) massive-MIMO system with short coherence time (CT). Five different physical correlation models are considered. The statistical structure of the massive-MIMO channel is exploited to find accurate CSI estimation. Training sequence design based on the transmit correlation matrix's eigenvectors is used to reduce estimation overhead. The article also examines the effect of channel hardening in the FDD massive-MIMO system.

PEERJ COMPUTER SCIENCE (2022)

添加到收藏夹

Article Energy & Fuels

Machine Learning Approach for Short-Term Load Forecasting Using Deep Neural Network

Majed A. Alotaibi

Summary: Power system demand forecasting is crucial for power system engineering, as planning and operation activities rely on accurate forecasting models. This paper compares the performance of three techniques - DNN, ANN, and DR - for short-term load forecasting. The results show that the DNN model outperforms the others.

ENERGIES (2022)

添加到收藏夹

Article Computer Science, Information Systems

Developing Novel Activation Functions Based Deep Learning LSTM for Classification

Mohamed H. Essai Ali et al.

Summary: This study proposes novel LSTM-based classifiers by developing the internal structure of LSTM neural networks using alternative state activation functions. The simulation results demonstrate that the proposed classifiers outperform the traditional tanh-based LSTM classifiers in terms of classification accuracy.

IEEE ACCESS (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Channel state information estimation for 5G wireless communication systems: recurrent neural networks approach

Mohamed Hassan Essai Ali et al.

Summary: A deep learning BiLSTM recurrent neural network-based channel state information estimator is proposed for 5G systems, showing superior performance with limited pilots and outperforming traditional estimators in symbol error rate and accuracy. Different classification layers, loss functions, and optimization algorithms are used for performance evaluation. The computational and training time complexities of the BiLSTM and LSTM estimators are provided, highlighting the promising potential of the deep learning approach for 5G and beyond communication systems.

PEERJ COMPUTER SCIENCE (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation

Davide Chicco et al.

Summary: Regression analysis is a key component of supervised machine learning, predicting continuous targets from other variables. While binary classification has targets with two values, regression can have multiple values. However, there is no consensus on a standard metric for evaluating regression results, with commonly used measures like MSE, RMSE, MAE, and MAPE having interpretability limitations.

PEERJ COMPUTER SCIENCE (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Efficiently inaccurate approximation of hyperbolic tangent used as transfer function in artificial neural networks

T. E. Simos et al.

Summary: By approximating tanh with specific cubic splines, we save computational costs in neural networks while sacrificing some accuracy in the final approximation, leading to improved performance.

NEURAL COMPUTING & APPLICATIONS (2021)

添加到收藏夹

Review Computer Science, Artificial Intelligence

A survey on modern trainable activation functions

Andrea Apicella et al.

Summary: In recent years, there has been a renewed interest in trainable activation functions, which can be trained during the learning process to improve neural network performance. Various models of trainable activation functions have been proposed in the literature, many of which are equivalent to adding neuron layers with fixed activation functions and simple local rules.

NEURAL NETWORKS (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Deep recurrent neural networks with word embeddings for Urdu named entity recognition

Wahab Khan et al.

ETRI JOURNAL (2020)

添加到收藏夹

Article Environmental Sciences

Comparison of Long Short Term Memory Networks and the Hydrological Model in Runoff Simulation

Hongxiang Fan et al.

WATER (2020)

添加到收藏夹

Review Mathematics, Interdisciplinary Applications

Machine-Learning Methods for Computational Science and Engineering

Michael Frank et al.

COMPUTATION (2020)

添加到收藏夹

Article Computer Science, Artificial Intelligence

A comparative performance analysis of different activation functions in LSTM networks for classification

Amir Farzad et al.

NEURAL COMPUTING & APPLICATIONS (2019)

添加到收藏夹

Review Computer Science, Information Systems

Review of Deep Learning Algorithms and Architectures

Ajay Shrestha et al.

IEEE ACCESS (2019)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Targeted style transfer using cycle consistent generative adversarial networks with quantitative analysis of different loss functions

Mannat Kaur et al.

INTERNATIONAL JOURNAL OF KNOWLEDGE-BASED AND INTELLIGENT ENGINEERING SYSTEMS (2018)

添加到收藏夹

Proceedings Paper Computer Science, Theory & Methods

A Survey on Deep Learning in Big Data

Mehdi Gheisari et al.

2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND ENGINEERING (CSE) AND IEEE/IFIP INTERNATIONAL CONFERENCE ON EMBEDDED AND UBIQUITOUS COMPUTING (EUC), VOL 2 (2017)

添加到收藏夹

Article Telecommunications

An Introduction to Deep Learning for the Physical Layer

Timothy O'Shea et al.

IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING (2017)

添加到收藏夹

Article Telecommunications

Deep Learning for Wireless Physical Layer: Opportunities and Challenges

Tianqi Wang et al.

CHINA COMMUNICATIONS (2017)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence

Denoising AutoEncoder in Neural Networks with Modified Elloitt Activation Function and Sparsity-Favoring Cost Function

Hasham Burhani et al.

3RD INTERNATIONAL CONFERENCE ON APPLIED COMPUTING AND INFORMATION TECHNOLOGY (ACIT 2015) 2ND INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND INTELLIGENCE (CSI 2015) (2015)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Bi-modal derivative activation function for sigmoidal feedforward networks

Sartaj Singh Sodhi et al.

NEUROCOMPUTING (2014)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Comparison of new activation functions in neural network for forecasting financial time series

Gecynalda S. da S. Gomes et al.

NEURAL COMPUTING & APPLICATIONS (2011)

添加到收藏夹

Article Computer Science, Artificial Intelligence

A Novel Connectionist System for Unconstrained Handwriting Recognition

Alex Graves et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2009)

添加到收藏夹

Article Computer Science, Artificial Intelligence

A case for the self-adaptation of activation functions in FFANNs

P Chandra et al.

NEUROCOMPUTING (2004)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.