Filters








84,647 Hits in 5.2 sec

Rewriting a Deep Generative Model [article]

David Bau, Steven Liu, Tongzhou Wang, Jun-Yan Zhu, Antonio Torralba
2020 arXiv   pre-print
In this paper, we introduce a new problem setting: manipulation of specific rules encoded by a deep generative model.  ...  A deep generative model such as a GAN learns to model a rich set of semantic and physical rules about the target distribution, but up to now, it has been obscure how such rules are encoded in the network  ...  We are grateful for the support of DARPA XAI FA8750-18-C-0004, DARPA SAIL-ON HR0011-20-C-0022, NSF 1524817 on Advancing Visual Recognition with Feature Visualizations, NSF BIGDATA 1447476, and a hardware  ... 
arXiv:2007.15646v1 fatcat:uye2azcsyjajhnkwy5tvigria4

Learning of human-like algebraic reasoning using deep feedforward neural networks

Cheng-Hao Cai, Yanyan Xu, Dengfeng Ke, Kaile Su
2018 Biologically Inspired Cognitive Architectures  
Briefly, in a reasoning system, a deep feedforward neural network is used to guide rewriting processes after learning from algebraic reasoning examples produced by humans.  ...  There is a wide gap between symbolic reasoning and deep learning. In this research, we explore the possibility of using deep learning to improve symbolic reasoning.  ...  Generally, rewriting requires a source expression s and a set of rewrite rules τ .  ... 
doi:10.1016/j.bica.2018.07.004 fatcat:tsevplqak5gcljvbkhwt2ix7ia

A Semantic Query Method for Deep Web

Hao Jiang, Wen Ju Liu, Li Li Lu
2013 Applied Mechanics and Materials  
Based on the idea of "functionality-centric", this paper proposes a complete set of oriented semantic query methods for Deep Web, builds up the relevant software architecture, provides a new method for  ...  full use of Deep Web data resources in semantic web environment through describing the establishment of the semantic environment, re-writing the SPARQL-to-SQL query, semantic packaging of semantic query  ...  appreciate our great thanks to those postgraduates fighting for their academic papers in figuring out precise data in the programs of the programs of "Jiangsu natural Science Foundation"(No.BK2008354 ), "Model  ... 
doi:10.4028/www.scientific.net/amm.347-350.2559 fatcat:p6dyiujb3vhmpmyv3sxwj2kzpa

Learning Generalizable Behavior via Visual Rewrite Rules [article]

Yiheng Xie, Mingxuan Li, Shangqun Yu, Michael Littman
2021 arXiv   pre-print
We also present preliminary results from a VRR agent that can explore, expand its rule set, and solve a game via planning with its learned VRR world model.  ...  In several classical games, our non-deep agent demonstrates superior performance, extreme sample efficiency, and robust generalization ability compared with several mainstream deep agents.  ...  Our VRR agent outperforms state-of-the-art model-free and model-based deep RL agents in tests of generalization and uses significantly fewer training samples.  ... 
arXiv:2112.05218v1 fatcat:vqry3op44rgkfklsz6irwjc4qe

Self-Supervised Learning to Prove Equivalence Between Programs via Semantics-Preserving Rewrite Rules [article]

Steve Kommrusch, Martin Monperrus, Louis-Noël Pouchet
2022 arXiv   pre-print
We propose a neural network architecture based on a transformer model to generate proofs of equivalence between program pairs.  ...  We represent programs using abstract syntax trees (AST), where a given set of semantics-preserving rewrite rules can be applied on a specific AST pattern to generate a transformed and semantically equivalent  ...  S4EQ: DEEP LEARNING TO FIND REWRITE RULE SEQUENCES We propose to use a deep learning model to find rewrite rule sequences which transform one program into a semantically equivalent target program.  ... 
arXiv:2109.10476v2 fatcat:5yijoz24ybep3jx6x3qqr7vnzq

DLUP: A Deep Learning Utility Prediction Scheme for Solid-State Fermentation Services in IIoT

Min Wang, Shanchen Pang, Tong Ding, Sibo Qiao, Xue Zhai, Shuo Wang, Neal N. Xiong, Zhengwen Huang
2022 IEEE Transactions on Industrial Informatics  
Furthermore, we propose a novel edge-rewritable Petri net to model the parameters collection and utility prediction of the SSF process and further verify their soundness.  ...  In this article, we propose a deep learning utility prediction (DLUP) scheme for the SSF in the Industrial Internet of Things, including parameters collection and utility prediction of the SSF process.  ...  Edge-Rewritable Petri Net Petri net is a workflow modeling and analysis tool.  ... 
doi:10.1109/tii.2021.3106590 fatcat:e5ltf75ihbhy5ecs2pxu7n25lm

Learning to Rewrite Queries

Yunlong He, Jiliang Tang, Hua Ouyang, Changsung Kang, Dawei Yin, Yi Chang
2016 Proceedings of the 25th ACM International on Conference on Information and Knowledge Management - CIKM '16  
In this paper, we propose a learning to rewrite framework that consists of a candidate generating phase and a candidate ranking phase.  ...  The task of query rewriting, aiming to alter a given query to a rewrite query that can close the gap and improve information retrieval performance, has attracted increasing attention in recent years.  ...  For example, since the SMT model aims at translating a sentence from a source language to a fluent and grammatically correct sentence in a target language, the SMT model prefers to rewrite the query "how  ... 
doi:10.1145/2983323.2983835 dblp:conf/cikm/HeTOKYC16 fatcat:nwvlype7yvcvhcat6guccu274i

A Haskell Hosted DSL for Writing Transformation Systems [chapter]

Andy Gill
2009 Lecture Notes in Computer Science  
It is intended for writing reasonably efficient rewrite systems, makes use of type families to provide a delimited generic mechanism for tree rewriting, and provides support for efficient identity rewrite  ...  KURE provides a small set of combinators that can be used to build parameterized term rewriting and general user-defined rule application.  ...  Acknowledgments I would like to thank the members of the Computer Systems Design Laboratory at the Information and Telecommunication Technology Center at the University of Kansas for providing a creative  ... 
doi:10.1007/978-3-642-03034-5_14 fatcat:6hjdl4p7traabmgqogu3natgmq

Towards Facilitating Empathic Conversations in Online Mental Health Support: A Reinforcement Learning Approach [article]

Ashish Sharma, Inna W. Lin, Adam S. Miner, David C. Atkins, Tim Althoff
2021 arXiv   pre-print
Our RL agent leverages a policy network, based on a transformer language model adapted from GPT-2, which performs the dual task of generating candidate empathic sentences and adding those sentences at  ...  We introduce a new task of empathic rewriting which aims to transform low-empathy conversational posts to higher empathy.  ...  ACKNOWLEDGMENTS We would like to thank TalkLife and Jamie Druitt for their support and for providing us access to a TalkLife dataset.  ... 
arXiv:2101.07714v3 fatcat:mfhbw5gryrfwbjw5usyaysfli4

D-PAGE: Diverse Paraphrase Generation [article]

Qiongkai Xu, Juyan Zhang, Lizhen Qu, Lexing Xie, Richard Nock
2018 arXiv   pre-print
We propose a simple method Diverse Paraphrase Generation (D-PAGE), which extends neural machine translation (NMT) models to support the generation of diverse paraphrases with implicit rewriting patterns  ...  In this paper, we investigate the diversity aspect of paraphrase generation. Prior deep learning models employ either decoding methods or add random input noise for varying outputs.  ...  The former methods can always be applied to a trained deep model.  ... 
arXiv:1808.04364v1 fatcat:umz4z5baknfhzkwlngj3pkzhci

Active Divergence with Generative Deep Learning – A Survey and Taxonomy [article]

Terence Broad, Sebastian Berns, Simon Colton, Mick Grierson
2021 arXiv   pre-print
use deep generative models in truly creative systems.  ...  Generative deep learning systems offer powerful tools for artefact generation, given their ability to model distributions of data and generate high-fidelity results.  ...  Targeted rewriting Bau et al. (2020) present a targeted approach to model rewriting.  ... 
arXiv:2107.05599v1 fatcat:vfsapuewi5btvbmqe2ehtsvr5m

Query Rewriting via Cycle-Consistent Translation for E-Commerce Search [article]

Yiming Qiu, Kang Zhang, Han Zhang, Songlin Wang, Sulong Xu, Yun Xiao, Bo Long, Wen-Yun Yang
2021 arXiv   pre-print
In this paper, we propose a novel deep neural network based approach to query rewriting, in order to tackle this problem.  ...  Then we introduce a novel cyclic consistent training algorithm in conjunction with state-of-the-art machine translation models to achieve the optimal performance in terms of query rewriting accuracy.  ...  Specifically, we develop a novel deep neural network model to query rewriting, which is composed of a cyclic translation formulation of query rewriting in Section III-B, a cyclic consistency likelihood  ... 
arXiv:2103.00800v1 fatcat:7ocr5lhcgzblpcmoca7qdpmncu

Arabic Word Segmentation with Long Short-Term Memory Neural Networks and Word Embedding

Abdulrahman Almuhareb, Waleed Alsanie, Abdulmohsen Al-Thubaity
2019 IEEE Access  
In this paper, we propose an Arabic word segmentation technique based on a bi-directional long short-term memory deep neural network.  ...  and Arabic word segmentation with the rewrite (more than 99% for frequent rewrite cases).  ...  Second, we consider a wider range of cases of segmentation rewriting (nine cases) than have been previously reported for Arabic word segmentation based on the output of the deep learning model without  ... 
doi:10.1109/access.2019.2893460 fatcat:h73mfiqyizefbkfhmths2ddw6q

Effective Loop Fusion in Polyhedral Compilation Using Fusion Conflict Graphs

Aravind Acharya, Uday Bondhugula, Albert Cohen
2020 ACM Transactions on Architecture and Code Optimization (TACO)  
In addition, DNNFusion includes 1) a novel mathematical-property-based graph rewriting framework to reduce evaluation costs and facilitate subsequent operator fusion, 2) an integrated fusion plan generation  ...  To achieve high accuracy, DNN models have become increasingly deep with hundreds or even thousands of operator layers, leading to high memory and computational requirements for inference.  ...  , 2) a novel mathematical-property-based graph rewriting, and 3) an integrated fusion plan generation.  ... 
doi:10.1145/3416510 fatcat:btm77em5izaknbi7tmxng7gpry

Semantic Role Labeling Guided Multi-turn Dialogue ReWriter [article]

Kun Xu and Haochen Tan and Linfeng Song and Han Wu and Haisong Zhang and Linqi Song and Dong Yu
2020 arXiv   pre-print
Experiments show that this information significantly improves a RoBERTa-based model that already outperforms previous state-of-the-art systems.  ...  For multi-turn dialogue rewriting, the capacity of effectively modeling the linguistic knowledge in dialog context and getting rid of the noises is essential to improve its performance.  ...  Our rewriting model is based on a pre-trained RoBERTa model that takes the outputs of SRL parsing and dialogue history as its inputs, before generating rewriting outputs word by word.  ... 
arXiv:2010.01417v1 fatcat:elsonjoyxvhbbcmymvkwrnamnu
« Previous Showing results 1 — 15 out of 84,647 results