Filters








28,686 Hits in 4.8 sec

Towards Similarity Graphs Constructed by Deep Reinforcement Learning [article]

Dmitry Baranchuk, Artem Babenko
2020 arXiv   pre-print
New algorithms for similarity graph construction are continuously being proposed and analyzed by both theoreticians and practitioners.  ...  Namely, we propose a probabilistic model of a similarity graph defined in terms of its edge probabilities and show how to learn these probabilities from data as a reinforcement learning task.  ...  Method In this section, we describe our approach for similarity graph construction based on reinforcement learning.  ... 
arXiv:1911.12122v2 fatcat:moev63syunf6no4hn2w2z46cdi

Evening the Score: Targeting SARS-CoV-2 Protease Inhibition in Graph Generative Models for Therapeutic Candidates [article]

Jenna Bilbrey, Logan Ward, Sutanay Choudhury, Neeraj Kumar, Ganesh Sivaraman
2021 arXiv   pre-print
Due to a sense of urgency, we chose well-validated models with unique strengths: an autoencoder that generates molecules with similar structures to a dataset of drugs with anti-SARS activity and a reinforcement  ...  learning algorithm that generates highly novel molecules.  ...  We next examine graph-based deep reinforcement learning (DQN) [Zhou et al. (2019) ] to generate candidates that are not constrained by their structural proximity to known anti-SARS Figure 1: Depiction  ... 
arXiv:2105.10489v1 fatcat:km7o3zdlzzdb5a4kavpstiisji

A Novel Deep Reinforcement Learning Based Stock Direction Prediction using Knowledge Graph and Community Aware Sentiments [article]

Anil Berk Altuner, Zeynep Hilal Kilimci
2021 arXiv   pre-print
learning side of the proposed model to construct the deep Q network.  ...  For this purpose, we firstly construct a social knowledge graph of users by analyzing relations between connections.  ...  Finally, a novel deep reinforcement learning based stock direction prediction model is constructed using knowledge graph and community aware sentiments.  ... 
arXiv:2107.00931v1 fatcat:c6oaw3y7nbayremuw2cxeekmmu

Fixed Priority Global Scheduling from a Deep Learning Perspective [article]

Hyunsung Lee, Michael Wang, Honguk Woo
2020 arXiv   pre-print
In this work, we first present how to adopt Deep Learning for real-time task scheduling through our preliminary work upon fixed priority global scheduling (FPGS) problems.  ...  We believe that there are many opportunities for leveraging advanced Deep Learning technologies to improve the quality of scheduling in various system configurations and problem scenarios.  ...  In ECC, the embedding for each node in a graph is learned by iteration over layers.  ... 
arXiv:2012.03002v2 fatcat:y5lx3ddctrfypebu7citrqm7ju

GraphOpt: Learning Optimization Models of Graph Formation [article]

Rakshit Trivedi, Jiachen Yang, Hongyuan Zha
2020 arXiv   pre-print
similar to those of the observed graph.  ...  GraphOpt poses link formation in graphs as a sequential decision-making process and solves it using maximum entropy inverse reinforcement learning algorithm.  ...  This work was supported by NSF IIS 1717916.  ... 
arXiv:2007.03619v1 fatcat:7raoevdcrrhx3ey7en6nedzeza

Strategies for Design of Molecular Structures with a Desired Pharmacophore Using Deep Reinforcement Learning

Atsushi Yoshimori, Enzo Kawasaki, Chisato Kanai, Tomohiko Tasaka
2020 Chemical and pharmaceutical bulletin  
In this study, we propose a computational strategy based on deep reinforcement learning for generating molecular structures with a desired pharmacophore.  ...  In addition, to extract selective molecules against a target protein, chemical genomics-based virtual screening (CGBVS) is used as post-processing method of deep reinforcement learning.  ...  This result shows that the Agent network learned the 3D pharmacophore information via deep reinforcement learning.  ... 
doi:10.1248/cpb.c19-00625 pmid:32115529 fatcat:jmyhojhg6vfqvomv24kdewcmgm

Joint Modeling of Dense and Incomplete Trajectories for Citywide Traffic Volume Inference

Xianfeng Tang, Boqing Gong, Yanwei Yu, Huaxiu Yao, Yandong Li, Haiyong Xie, Xiaoyu Wang
2019 The World Wide Web Conference on - WWW '19  
Our approach employs a high-fidelity traffic simulator and deep reinforcement learning to recover full vehicle movements from the incomplete trajectories.  ...  In order to jointly model the recovered trajectories and dense GPS trajectories, we construct spatiotemporal graphs and use multi-view graph embedding to encode the multi-hop correlations between road  ...  Deep Reinforcement Learning for Parameter Tuning.  ... 
doi:10.1145/3308558.3313621 dblp:conf/www/TangGYYL0W19 fatcat:qnume2ocarb6bmkksxtnmlqhia

Using Cyber Terrain in Reinforcement Learning for Penetration Testing [article]

Rohit Gangupantulu, Tyler Cody, Paul Park, Abdul Rahman, Logan Eisenbeiser, Dan Radke, Ryan Clark
2021 arXiv   pre-print
Reinforcement learning (RL) has been applied to attack graphs for penetration testing, however, trained agents do not reflect reality because the attack graphs lack operational nuances typically captured  ...  In particular, current practice constructs attack graphs exclusively using the Common Vulnerability Scoring System (CVSS) and its components.  ...  Reinforcement learning for penetration testing uses the attack graph model [2] - [8] .  ... 
arXiv:2108.07124v1 fatcat:wl5l3durhzc4bdvs5l454ude7i

New Ideas and Trends in Deep Multimodal Content Understanding: A Review [article]

Wei Chen and Weiping Wang and Li Liu and Michael S. Lew
2020 arXiv   pre-print
The focus of this survey is on the analysis of two modalities of multimodal deep learning: image and text.  ...  Unlike classic reviews of deep learning where monomodal image classifiers such as VGG, ResNet and Inception module are central topics, this paper will examine recent multimodal deep models and structures  ...  Acknowledgments This work was supported by LIACS MediaLab at Leiden University and China Scholarship Council (CSC No. 201703170183). We appreciate the helpful editing work from Dr. Erwin Bakker.  ... 
arXiv:2010.08189v1 fatcat:2l7molbcn5hf3oyhe3l52tdwra

A Survey of Deep Reinforcement Learning in Recommender Systems: A Systematic Review and Future Directions [article]

Xiaocong Chen, Lina Yao, Julian McAuley, Guanglin Zhou, Xianzhi Wang
2021 arXiv   pre-print
of the recent trends of deep reinforcement learning in recommender systems.  ...  In light of the emergence of deep reinforcement learning (DRL) in recommender systems research and several fruitful results in recent years, this survey aims to provide a timely and comprehensive overview  ...  In contrast, deep reinforcement learning (DRL) aims to train an agent that can learn from interaction trajectories provided by the environment by combining the power of deep learning and reinforcement  ... 
arXiv:2109.03540v2 fatcat:5gwrbfcj3rc7jfkd54eseck5ga

Towards personalized human AI interaction - adapting the behavior of AI agents using neural signatures of subjective interest [article]

Victor Shih, David C Jangraw, Paul Sajda, Sameer Saproo
2017 arXiv   pre-print
Reinforcement Learning AI commonly uses reward/penalty signals that are objective and explicit in an environment -- e.g. game score, completion time, etc. -- in order to learn the optimal strategy for  ...  Here, we show how a hybrid brain-computer-interface (hBCI), which detects an individual's level of interest in objects/events in a virtual environment, can be used to adapt the behavior of a Deep Reinforcement  ...  ACKNOWLEDGMENT The work was partially funded by the Army Research Laboratory under Cooperative agreement number W911NF-10-2-0022. This research was partially supported by BRAIQ, Inc.  ... 
arXiv:1709.04574v1 fatcat:eud2yqbsmnbvvpmiw7htwrs3va

Deep Learning and Knowledge-Based Methods for Computer Aided Molecular Design – Toward a Unified Approach: State-of-the-Art and Future Directions [article]

Abdulelah S. Alshehri, Rafiqul Gani, Fengqi You
2020 arXiv   pre-print
In view of the computational challenges plaguing knowledge-based methods and techniques, we survey the current state-of-the-art applications of deep learning to molecular design as a fertile approach towards  ...  The main focus of the survey is given to deep generative modeling of molecules under various deep learning architectures and different molecular representations.  ...  deep learning architectures.  ... 
arXiv:2005.08968v2 fatcat:p2mfdbkqsjekxa6qwoz5xpfzuu

Vulcan: Solving the Steiner Tree Problem with Graph Neural Networks and Deep Reinforcement Learning [article]

Haizhou Du and Zong Yan and Qiao Xiang and Qinqing Zhan
2021 arXiv   pre-print
To this end, we design a novel model Vulcan based on novel graph neural networks and deep reinforcement learning.  ...  Given an STP instance, Vulcan uses this embedding to encode its pathrelated information and sends the encoded graph to a deep reinforcement learning component based on a double deep Q network (DDQN) to  ...  We see the presented work as a step towards a new family of solvers for NP-hard problems that leverage both deep learning and reinforcement learning.  ... 
arXiv:2111.10810v1 fatcat:hiphlo2pojbd7jdvdweo25v764

SeaPearl: A Constraint Programming Solver guided by Reinforcement Learning [article]

Félix Chalumeau
2021 arXiv   pre-print
This paper presents the proof of concept for SeaPearl, a new CP solver implemented in Julia, that supports machine learning routines in order to learn branching decisions using reinforcement learning.  ...  Recent works have shown that reinforcement learning can be successfully used for driving the search phase of constraint programming (CP) solvers.  ...  In this paper, we propose a flexible and open-source research framework towards the hybridization of constraint programming and deep reinforcement learning.  ... 
arXiv:2102.09193v2 fatcat:gunjyfdi2jbkjk6enzhn6wq6su

Reinforcement Learning for Combinatorial Optimization: A Survey [article]

Nina Mazyavkina and Sergey Sviridov and Sergei Ivanov and Evgeny Burnaev
2020 arXiv   pre-print
Reinforcement learning (RL) proposes a good alternative to automate the search of these heuristics by training an agent in a supervised or self-supervised manner.  ...  Many traditional algorithms for solving combinatorial optimization problems involve using hand-crafted heuristics that sequentially construct a solution.  ...  This work on DQN has given rise to the whole field of Deep Reinforcement Learning methods.  ... 
arXiv:2003.03600v3 fatcat:ofc6gzf2fzhchjawbxfiin3354
« Previous Showing results 1 — 15 out of 28,686 results