scholarly journals Index selection for NoSQL database with deep reinforcement learning

2021 ◽  
Vol 561 ◽  
pp. 20-30
Author(s):  
Yu Yan ◽  
Shun Yao ◽  
Hongzhi Wang ◽  
Meng Gao
Keyword(s):  
Reinforcement Learning ◽  
Index Selection ◽  
Nosql Database ◽  
Selection For

Deep reinforcement learning based multicast mode selection for SFN

2021 ◽  
Author(s):  
Hao Cheng ◽  
Yihang Huang ◽  
Dazhi He ◽  
Yin Xu ◽  
Yanfeng Wang ◽  
...  
Keyword(s):  
Reinforcement Learning ◽  
Mode Selection ◽  
Selection For

Index selection for OLAP

2002 ◽  
Author(s):  
H. Gupta ◽  
V. Harinarayan ◽  
A. Rajaraman ◽  
J.D. Ullman
Keyword(s):  
Index Selection ◽  
Selection For

scholarly journals Querying NoSQL with Deep Learning to Answer Natural Language Questions

2019 ◽  
Vol 33 ◽  
pp. 9416-9421
Author(s):  
Sebastian Blank ◽  
Florian Wilhelm ◽  
Hans-Peter Zorn ◽  
Achim Rettinger
Keyword(s):  
Deep Learning ◽  
Reinforcement Learning ◽  
Natural Language ◽  
Question Answering ◽  
Query Languages ◽  
Domain Specific ◽  
Nosql Database ◽  
End To End ◽  
Database Operations ◽  
Almost All

Almost all of today’s knowledge is stored in databases and thus can only be accessed with the help of domain specific query languages, strongly limiting the number of people which can access the data. In this work, we demonstrate an end-to-end trainable question answering (QA) system that allows a user to query an external NoSQL database by using natural language. A major challenge of such a system is the non-differentiability of database operations which we overcome by applying policy-based reinforcement learning. We evaluate our approach on Facebook’s bAbI Movie Dialog dataset and achieve a competitive score of 84.2% compared to several benchmark models. We conclude that our approach excels with regard to real-world scenarios where knowledge resides in external databases and intermediate labels are too costly to gather for non-end-to-end trainable QA systems.

Effect of index selection for reduced backfat thickness and increased growth rate on sow weight changes through two parities in swine

1992 ◽  
Vol 72 (2) ◽  
pp. 403-408 ◽  
Author(s):  
R. M. McKay
Keyword(s):  
Growth Rate ◽  
Key Words ◽  
Selection Pressure ◽  
Backfat Thickness ◽  
Index Selection ◽  
Weight Changes ◽  
Selection For ◽  
The Impact

The impact of continued selection for reduced backfat thickness and improved growth rate on weight changes in Yorkshire and Hampshire sows, from breeding as gilts to weaning their second litter, was studied. Results suggest that some sow weight changes may have been influenced by selection pressure exerted. Key words: Sows, Canadian Yorkshire, Hampshire, liveweight change

Index selection for improved growth and carcass characteristics in a population of Large White pigs

1988 ◽  
Vol 46 (2) ◽  
pp. 265-275 ◽  
Author(s):  
M. Ellis ◽  
J. P. Chadwick ◽  
W. C. Smith ◽  
R. Laird
Keyword(s):  
Carcass Characteristics ◽  
Live Weight ◽  
Conversion Ratio ◽  
Index Selection ◽  
Food Conversion ◽  
Genetic Progress ◽  
Large White ◽  
Selection For ◽  
On Farm ◽  
Ad Libitum Feeding

AbstractAn experiment was carried out over 11 years to investigate selection for economy of production and carcass lean content under ad libitum feeding in Large White pigs. Two lines, a selection (S) and a control (C) line, were involved in the study. The S line comprised 80 females and 10 males and was based at two centres. Boars were performance tested centrally at one of the centres and gilts were on-farm tested. Testing was carried out in groups of two or three full-sibs over the live-weight range 27 o t 87 kg. Selection was based on an index (I) incorporating individual daily live-weight gain (DLWG) and ultrasonically measured backfat thickness (USBF) and a group food conversion ratio (FCR) and generations were overlapping. The C line (32 females and 16 males) was maintained at one centre and males were performance tested alongside S boars to monitor genetic progress. Cumulative realized selection differentials over years 1 to 10 were equivalent to 5·5, 51, 7·0 and 9·4 phenotypic standard deviations for DLWG, USBF, FCR and I respectively and generation intervals averaged 17·0 months. There was little genetic change in DLWG, however, USBF, FCR and I showed substantial improvements with cumulative responses in year 11 of —12·3 mm, —0·22 kg/kg and +45·2 points respectively. The reduction in USBF occurred in the first half of the study with no further improvements being achieved after year 6. This study illustrates the effectiveness of index selection for a limited number of economically important traits but highlights limitations to this approach.

scholarly journals Selection Indices for Simultaneous Selection for Pod Yield and Leafspot Resistance in Peanut (Arachis hypogaea L.)1

1987 ◽  
Vol 14 (1) ◽  
pp. 51-54 ◽  
Author(s):  
R. N. Iroume ◽  
D. A. Knauft
Keyword(s):  
Arachis Hypogaea ◽  
Breeding Value ◽  
Direct Selection ◽  
Index Selection ◽  
Cercospora Arachidicola ◽  
Pod Yield ◽  
Arachis Hypogaea L ◽  
Yield Gain ◽  
Selection For ◽  
Disease Pressure

Abstract The objective of this study was to investigate the possibility of index selection for pod yield and leafspot resistance, causal organisms Cercospora arachidicola Hori and Cercosporidium personatum (Berk. and Curt.) Deighton, in peanut (Arachis hypogaea L.). Eleven crosses were evaluated in the S1 and S2 generations under natural disease pressure for pod yield and disease reaction. Leafspot severity was measured by leaf necrotic area and defoliation. Variances for each trait and covariances between traits were estimated by progeny analysis. Economic weights assigned to the different characters were derived as the ratio of the pod yield for a genotype to the theoretical pod yield gain from leafspot resistance. Indices were constructed and their efficiency both as predictor of the breeding value and relative to individual selection for either trait was evaluated. Index selection for increased pod yield and reduced leafspot susceptibility was between 87% and 100% as efficient as direct selection for either trait. Combining several leafspot severity readings with pod yield did not improve the efficiency of the index. Fourteen to 17% of the breeding value of the population was accounted for by the traits combined in various indices. This indicated that there is potential for improvement of these indices.

Sign in / Sign up

Export Citation Format

Share Document