903 Hits in 4.4 sec

Learning Efficient Algorithms with Hierarchical Attentive Memory [article]

Marcin Andrychowicz, Karol Kurach
2016 arXiv   pre-print
We show that an LSTM network augmented with HAM can learn algorithms for problems like merging, sorting or binary searching from pure input-output examples.  ...  It is based on a binary tree with leaves corresponding to memory cells.  ...  Acknowledgements We would like to thank Nando de Freitas, Alexander Graves, Serkan Cabi, Misha Denil and Jonathan Hunt for helpful comments and discussions.  ... 
arXiv:1602.03218v2 fatcat:cs23omiufnarrecwtaumfcz7xi

Jointly Learning Sentence Embeddings and Syntax with Unsupervised Tree-LSTMs [article]

Jean Maillard, Stephen Clark, Dani Yogatama
2017 arXiv   pre-print
Our model simultaneously optimises both the composition function and the parser, thus eliminating the need for externally-provided parse trees which are normally required for Tree-LSTM.  ...  It can therefore be seen as a tree-based RNN that is unsupervised with respect to the parse trees.  ...  The authors use reinforcement learning to learn tree structures for a neural network model similar to Bowman et al. (2016) , taking performance on a downstream task that uses the computed sentence representations  ... 
arXiv:1705.09189v1 fatcat:p6st2uaddvg2dimsp2u3hk52vm

Enhancement of Natural Language to SQL Query Conversion using Machine Learning Techniques

Akshar Prasad, Sourabh S, Yashwanth YS, Shetty Rohan, Shobha G, Deepamala N
2020 International Journal of Advanced Computer Science and Applications  
The major challenges faced by a user accessing this data is to learn the querying language and understand the various syntax associated with it.  ...  Query given in the form of Natural Language helps any naïve user to access database without learning the query languages.  ...  ) = 'australia' show subtotal of orders for helmet show subtotal of orders for ProductSubCategoryName 'helmet' SELECT SUM( t_saldtls.SalesOrderint ) FROM t_prds INNER JOIN t_saldtls ON t_prds.ProductKey  ... 
doi:10.14569/ijacsa.2020.0111260 fatcat:ze6htssqhvcttg4c5vpfryp4nq

Jointly learning sentence embeddings and syntax with unsupervised Tree-LSTMs

Jean Maillard, Stephen Clark, Dani Yogatama
2019 Natural Language Engineering  
The models simultaneously optimise both the composition function and the parser, thus eliminating the need for externally provided parse trees, which are normally required for Tree-LSTMs.  ...  They can therefore be seen as tree-based recurrent neural networks that are unsupervised with respect to the parse trees.  ...  The authors use reinforcement learning to learn tree structures for a neural network model similar to Bowman et al. (2016) , taking performance on a downstream task that uses the computed sentence representations  ... 
doi:10.1017/s1351324919000184 fatcat:7scvo7uz2rchfpij3kvy7w24ru

Using deep reinforcement learning approach for solving the multiple sequence alignment problem

Reza Jafari, Mohammad Masoud Javidi, Marjan Kuchaki Rafsanjani
2019 SN Applied Sciences  
In the present paper, we use a deep reinforcement learning (DRL) approach for solving the multiple sequence alignment problem which is an NP-complete problem.  ...  Furthermore, the actor-critic algorithm with experience-replay method is used for much quicker convergence process.  ...  In this paper, we apply the deep reinforcement learning (DRL) [12] approach for the MSA problem.  ... 
doi:10.1007/s42452-019-0611-4 fatcat:x2xxdesyubb5ne6nljh3a4lp7m

A Unified Transferable Model for ML-Enhanced DBMS [article]

Ziniu Wu, Pei Yu, Peilun Yang, Rong Zhu, Yuxing Han, Yaliang Li, Defu Lian, Kai Zeng, Jingren Zhou
2021 arXiv   pre-print
Recently, the database management system (DBMS) community has witnessed the power of machine learning (ML) solutions for DBMS tasks.  ...  Moreover, for each retraining, they require an excessive amount of training data, which is very expensive to acquire and unavailable for a new DB.  ...  ML-based CostEst methods use treebased models (such as tree convolution [23] and tree-LSTM [31] ) to encode a plan and map the encoding to its estimated costs. • Join order selection (JoinSel) decides  ... 
arXiv:2105.02418v3 fatcat:ljb66dxlkvhdtbwnam73e5mxm4

A BiLSTM cardinality estimator in complex database systems based on attention mechanism

Qiang Zhou, Guoping Yang, Haiquan Song, Jin Guo, Yadong Zhang, Shengjie Wei, Lulu Qu, Louis Alberto Gutierrez, Shaojie Qiao
2021 CAAI Transactions on Intelligence Technology  
The results show that the deep learning model can significantly improve the quality of cardinality estimation, which is a vital role in query optimisation for complex databases.  ...  Finally, the BiLSTM network and attention mechanism are employed to deal with word vectors.  ...  For the selection of join order, this kind of problem needs to constantly interact with the intermediate results of the current query.  ... 
doi:10.1049/cit2.12069 fatcat:gltq7un7gfa2feuo6nc5w26bua

Progressive Neural Index Search for Database System [article]

Sai Wu, Xinyi Yu, Xiaojie Feng, Feifei Li, Wei Cao, Gang Chen
2020 arXiv   pre-print
We achieve the two goals for a given workload and dataset with one RNN-powered reinforcement learning model.  ...  Ordered blocks are implemented using B+-tree nodes or skip lists, while unordered blocks adopt hash functions with different configurations.  ...  The learning model tries dierent join orders during dierent time slices and promising plan is selected. Neo [13] , on the other hand, tries to rewrite the database optimizer in a learning language.  ... 
arXiv:1912.07001v2 fatcat:u3k7gorbyvhnfoolrf3hidyrcy

Energy Consumption Forecasting in Korea Using Machine Learning Algorithms

Sun-Youn Shin, Han-Gyun Woo
2022 Energies  
is less unknown irregularity in the time series, but machine learning can work better with unexpected irregular time series data.  ...  To bridge this gap, this paper compared three different machine learning algorithms, namely the Random Forest (RF) model, XGBoost (XGB) model, and Long Short-Term Memory (LSTM) model.  ...  In order to analyze the LSTM Model As was with the RF model and the XGBoost model, the ANN model LSTM utilized the same data.  ... 
doi:10.3390/en15134880 fatcat:qqqj4t6apjgitgpnf55tqn7mie

A Survey of Machine Learning-Based System Performance Optimization Techniques

Hyejeong Choi, Sejin Park
2021 Applied Sciences  
This paper reviews 11 machine learning-based system performance optimization approaches from nine recent papers based on well-known machine learning models such as perceptron, LSTM, and RNN.  ...  The result shows that machine learning-based system performance optimization has an important potential for future research.  ...  Because RMI trains only for each range of data, such as B-Tree, RMI can easily build with a simple model. In addition, this paper supports B-Tree.  ... 
doi:10.3390/app11073235 fatcat:fjjguf3x6fdjtm543iejmr255i

A View on Deep Reinforcement Learning in System Optimization [article]

Ameer Haj-Ali, Nesreen K. Ahmed, Ted Willke, Joseph Gonzalez, Krste Asanovic, Ion Stoica
2019 arXiv   pre-print
We conclude with a discussion on open challenges and potential directions for pushing further the integration of reinforcement learning in system optimization.  ...  reinforcement learning.  ...  Feature selection and policy optimization for distributed instruction placement using reinforcement learning.  ... 
arXiv:1908.01275v3 fatcat:ih52psaazzcs3pulz4nnnjk2di

Recent Advances in SQL Query Generation: A Survey [article]

Jovan Kalajdjieski, Martina Toshevska, Frosina Stojanovska
2020 arXiv   pre-print
We describe models with various architectures such as convolutional neural networks, recurrent neural networks, pointer networks, reinforcement learning, etc.  ...  With the rise of deep learning techniques, there is extensive ongoing research in designing a suitable natural language interface to relational databases.  ...  The model is trained in different ways: with standard crossentropy loss over the pairs of question and SQL query, and with reinforcement learning with policy gradient as in [15] . F.  ... 
arXiv:2005.07667v1 fatcat:ilxq5lc4yvfz7lmtb5kedtrdzy

Thread Popularity Prediction and Tracking with a Permutation-invariant Model

Hou Pong Chan, Irwin King
2018 Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing  
This task has been formulated as a reinforcement learning problem, in which the reward of the agent is the sum of positive responses received by the recommended comments.  ...  We would like to thank Jiani Zhang, Hongyi Zhang, Wang Chen, Yifan Gao, and Xiaotian Yu for their comments.  ...  Related Work Reinforcement Learning in Text-based Tasks Reinforcement learning has been widely applied in various text-based tasks.  ... 
doi:10.18653/v1/d18-1376 dblp:conf/emnlp/ChanK18 fatcat:gyzfrzbfajcrjex3lufryx5f5q

Deep Neural Network Model for Proficient Crop Yield Prediction

K. Pravallika, G. Karuna, K. Anuradha, V. Srilakshmi, S. Tummala, S. Kosaraju, P. Bobba, S. Singh
2021 E3S Web of Conferences  
Forecasting crop yield well before harvest time can help farmers for selling and storage. Agriculture deals with large datasets and knowledge process.  ...  Based on the study of various survey papers it has been found that in all the crop predictions, various deep learning, machine learning and ANN algorithms implemented to predict yield forecast and the  ...  For text, speech and general-time series LSTMs are good for such sequences of data. Deep Neural Networks For association of input and outputs deep learning algorithms are used along with networks.  ... 
doi:10.1051/e3sconf/202130901031 fatcat:72gvp3szzfeqfaw5iwoaplnjg4

QueryFormer: A Tree Transformer Model for Query Plan Representation

Yue Zhao, Gao Cong, Jiachen Shi, Chunyan Miao
2022 Proceedings of the VLDB Endowment  
Machine learning has become a prominent method in many database optimization problems such as cost estimation, index selection and query optimization.  ...  In addition, to effectively capture the information flow following the tree structure of a query plan, we develop a tree-structured model with the attention mechanism.  ...  Specifically, it takes a pair of physical query plans as input, and predicts which plan is better. (2) For join order selection, ReJOIN [18] and RTOS [36] use reinforcement learning to determine the  ... 
dblp:journals/pvldb/ZhaoCSM22 fatcat:tip53qrdlrafxnm3srmgdzaydm
« Previous Showing results 1 — 15 out of 903 results