Policy gradient empowered LSTM with dynamic skips for irregular time series data

插补(统计学) 缺少数据 计算机科学 强化学习 人工智能 时间序列 回归 机器学习 数据挖掘 系列(地层学) 统计 数学 生物 古生物学
作者
Philip B. Weerakody,Kok Wai Wong,Guanjin Wang
出处
期刊:Applied Soft Computing [Elsevier BV]
卷期号:142: 110314-110314 被引量:8
标识
DOI:10.1016/j.asoc.2023.110314
摘要

Time series modelling has been successfully handled by Long Short-Term Memory (LSTM) models. Yet their performance can be severely inhibited by the occurrence of missing values prevalent in many real-life datasets. Many previous studies have been dedicated to imputation methods for generating a complete time series sequence, which have their limitations in terms of imputation bias and inaccuracies. In this paper, we propose a new LSTM model incorporating policy gradient (PG) based reinforcement learning called PG-LSTM, which can mitigate the effect of missing data and capture time-based input feature patterns more effectively to improve prediction performance. Inspired by numerous sequence models' successes in improving the efficiency of processing language data by skipping irrelevant tokens, the PG-LSTM introduces dynamic skip connections between LSTM cell states for time series data for classification and regression tasks for the first time. Specifically, the proposed model comprises a modified LSTM cell architecture that can internally call a policy-based reinforcement learning agent to generate a skipping action, allowing the model to dynamically select the optimal subset of hidden and cell states from past states to capture periodic and non-periodic patterns within a time series sequence. Moreover, the PG-LSTM also designs a lightweight imputation layer using a simple missing value imputation strategy while incorporating missing indicators and skipping segments of unimportant data to reduce the limitations associated with imputed data for handling missing values. Our experimental results on regression and classification tasks on time series data with high rates of missing values demonstrate that the PG-LSTM improves performance against current gated recurrent neural networks (RNN) and conventional non-neural network algorithms. The PG-LSTM can enhance AUC by up to 18.5% in the classification task and RMSE by up to 19.3% in the regression task over gated RNN models, respectively. Our findings are also statistically analysed using statistical significance testing with post hoc analysis.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
聪明水之完成签到,获得积分10
2秒前
Xiaoqiu完成签到 ,获得积分10
2秒前
外向访卉发布了新的文献求助10
2秒前
所所应助木木彡采纳,获得10
3秒前
5秒前
顺顺尼发布了新的文献求助10
5秒前
忐忑的鞅发布了新的文献求助10
7秒前
7秒前
英姑应助orz采纳,获得10
7秒前
8秒前
11秒前
董冬咚发布了新的文献求助10
11秒前
12秒前
茸茸茸发布了新的文献求助10
14秒前
科研通AI2S应助左肩微笑采纳,获得10
14秒前
15秒前
15秒前
wl1700发布了新的文献求助10
16秒前
瘦瘦的枫叶完成签到 ,获得积分10
16秒前
16秒前
17秒前
liuzhanyu发布了新的文献求助10
18秒前
18秒前
晓湫发布了新的文献求助10
19秒前
茸茸茸完成签到,获得积分10
19秒前
20秒前
琲珂发布了新的文献求助10
20秒前
21秒前
白樱恋曲发布了新的文献求助10
21秒前
22秒前
今后应助但愿人长久采纳,获得10
22秒前
你你你发布了新的文献求助10
23秒前
iu发布了新的文献求助10
23秒前
可爱的小桃完成签到,获得积分10
24秒前
hh发布了新的文献求助10
24秒前
25秒前
莫小烦完成签到,获得积分10
27秒前
忐忑的鞅完成签到,获得积分10
28秒前
29秒前
BlingBling完成签到,获得积分10
29秒前
高分求助中
The Mother of All Tableaux Order, Equivalence, and Geometry in the Large-scale Structure of Optimality Theory 2400
Ophthalmic Equipment Market by Devices(surgical: vitreorentinal,IOLs,OVDs,contact lens,RGP lens,backflush,diagnostic&monitoring:OCT,actorefractor,keratometer,tonometer,ophthalmoscpe,OVD), End User,Buying Criteria-Global Forecast to2029 2000
Optimal Transport: A Comprehensive Introduction to Modeling, Analysis, Simulation, Applications 800
Official Methods of Analysis of AOAC INTERNATIONAL 600
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 588
T/CIET 1202-2025 可吸收再生氧化纤维素止血材料 500
Comparison of adverse drug reactions of heparin and its derivates in the European Economic Area based on data from EudraVigilance between 2017 and 2021 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3953141
求助须知:如何正确求助?哪些是违规求助? 3498499
关于积分的说明 11092224
捐赠科研通 3229097
什么是DOI,文献DOI怎么找? 1785211
邀请新用户注册赠送积分活动 869255
科研通“疑难数据库(出版商)”最低求助积分说明 801415