5,751 Hits in 7.3 sec

Near-Optimal Multi-Perturbation Experimental Design for Causal Structure Learning [article]

Scott Sussex
2021 arXiv   pre-print
Causal structure learning is a key problem in many domains. Causal structures can be learnt by performing experiments on the system of interest.  ...  In this paper, we develop efficient algorithms for optimizing different objective functions quantifying the informativeness of a budget-constrained batch of experiments.  ...  Interventions We use the term intervention to refer to a set I ⊂ [p] of perturbation targets (variables).  ... 
arXiv:2105.14024v2 fatcat:qktahxzzzresdersc6gwjtcdfy

Improving human understanding and design of complex multi-level systems with animation and parametric relationship supports

Paul Egan, Christian Schunn, Jonathan Cagan, Philip LeDuc
2015 Design Science  
Here, we investigate engineering students' capacity to search for optimal nanoscale biosystem designs with stochastic component and system behaviors.  ...  with agent-based animations that emphasized inter-level causality learning.  ...  Financial support Partial funding for this study was provided by the National Defense Science and Engineering Graduate Fellowship and by the National Science Foundation under grant CMMI-1160840.  ... 
doi:10.1017/dsj.2015.3 fatcat:ntqxkbtvhbbqllte3qif7m5lzi

Subject index

2021 Journal of Systems Engineering and Electronics  
•••••••••••••••••••••••••••••••••••• 3-658 Causal constraint pruning for exact learning of Bayesian network structure •••••••••••••••••••••••••••••••••••• 4-854 Range-spread target detector via coherent  ...  •••••••••••••••••••••••• 2-380 RFC: a feature selection algorithm for software defect prediction ••••••••••••••••••••••••••••••••••••••••••••••••••••• 2-389 A pilot allocation method for multi-cell multi-user  ...  An integrated simulation system for operating solar sail spacecraft An iterated local coordinate-exchange algorithm for constructing experimental designs for multi-dimensional constrained spaces ADC-GERT  ... 
doi:10.23919/jsee.2021.9679721 fatcat:x76rw4j6bjbcnamwooh37hozhq

Towards Robust Representation of Limit Orders Books for Deep Learning Models [article]

Yufei Wu, Mahmoud Mahfouz, Daniele Magazzeni, Manuela Veloso
2021 arXiv   pre-print
We analyse the issues associated with the commonly-used LOB representation for machine learning models from both theoretical and experimental perspectives.  ...  The lack of attention on the robustness of representations may boost risks when using data-driven machine learning models for trading in the financial markets.  ...  This paper was prepared for informational purposes in part by the Artificial Intelligence Research group of JPMorgan Chase & Co and its affiliates ("J.P.  ... 
arXiv:2110.05479v1 fatcat:ff3lzlq2hrhkbgkojc7tpvvcty

A Simulation-Based Test of Identifiability for Bayesian Causal Inference [article]

Sam Witty, David Jensen, Vikash Mansinghka
2021 arXiv   pre-print
This paper introduces a procedure for testing the identifiability of Bayesian models for causal inference.  ...  This approach expresses causal assumptions as priors over functions in a structural causal model, including flexible priors using Gaussian processes.  ...  Simulation-based identifiability is fully automated for any prior over structural causal models with a differentiable likelihood and causal effect, covering quasi-experimental designs that previously required  ... 
arXiv:2102.11761v1 fatcat:k5qyprfkerft3k4znd22hcdlfq

Recurrent neural circuits overcome partial inactivation by compensation and re-learning [article]

Colin J Bredenberg, Cristina Savin, Roozbeh Kiani
2021 bioRxiv   pre-print
However, complexities of neural circuits challenge interpretation of experimental results, necessitating theoretical frameworks for systematic explorations.  ...  Finally, networks that can "learn" during inactivation recover function quickly, often much faster than the original training time.  ...  Shushruth for inspiring discussions and feedback on earlier versions of the manuscript. We thank Owen Marschall for sharing code to implement the RFLO algorithm.  ... 
doi:10.1101/2021.11.12.468273 fatcat:wz6szqkdqvc4dnn4kiqcg6rpjm

Long-range Event-level Prediction and Response Simulation for Urban Crime and Global Terrorism with Granger Networks [article]

Timmy Li, Yi Huang, James Evans, Ishanu Chattopadhyay
2019 arXiv   pre-print
We conclude that while crime operates near an equilibrium quickly dissipating perturbations, terrorism does not.  ...  Standard machine learning approaches are promising, but lack interpretability, are generally interpolative, and ineffective for precise future interventions with costly and wasteful false positives.  ...  ACKNOWLEDGMENTS Our work greatly benefited from discussion of everyone who participated in our workshop series on crime prediction at the Neubauer Collegium for culture and society 63 , and with those  ... 
arXiv:1911.05647v1 fatcat:64vjy6jvjzdnbmp5mh6icrjb3a

Knowledge-guided artificial intelligence technologies for decoding complex multiomics interactions in cells

Dohoon Lee, Sun Kim
2021 Clinical and experimental pediatrics  
Artificial intelligence (AI) technologies, including deep learning models, are optimal choices for handling complex nonlinear relationships between features that are scalable and produce large amounts  ...  Thus, to facilitate further development of knowledge-guided AI technologies for the modeling of multi-omics interactions, here we review representative bioinformatics applications of deep learning models  ...  The most recent and prominent example emphasizing the power of a well-designed neural network architecture is AlphaFold2 8) , which predicts the structure of a protein with near-experimental accuracy  ... 
doi:10.3345/cep.2021.01438 pmid:34844399 pmcid:PMC9082244 fatcat:7f7mcd2bazdaxfcnnhu77fxxji

Special Issue "Complex Dynamic System Modelling, Identification and Control"

Quanmin Zhu, Giuseppe Fusco, Jing Na, Weicun Zhang, Ahmad Taher Azar
2022 Entropy  
Systems are naturally or purposely formed with functional components and connection structures [...]  ...  Therefore, the guest editors hope that the readers can benefit from these published articles, the meaningful concepts and insights presented, emerging techniques, and inspiration for their future research  ...  Aiming to address this problem, this paper combines the particle swarm optimization (PSO) with the coefficient diagram method (CDM) and proposes a robust controller design strategy for the MIMO systems  ... 
doi:10.3390/e24030380 pmid:35327891 pmcid:PMC8947381 fatcat:ujmscwlup5a4feuo2kl7wivr44

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
This survey serves as introductory material for readers from academia and industry into the topic and identifies notable opportunities for further research.  ...  of the recent trends of deep reinforcement learning in recommender systems.  ...  Hence, DeepChain designs a multi-agent setting that adopts several agents to learn consecutive scenarios and jointly optimizes multiple recommendation policies.  ... 
arXiv:2109.03540v2 fatcat:5gwrbfcj3rc7jfkd54eseck5ga

FIND:Explainable Framework for Meta-learning [article]

Xinyue Shao, Hongzhi Wang, Xiao Zhu, Feng Xiong
2022 arXiv   pre-print
Since the traditional meta-learning technique lacks explainability, as well as shortcomings in terms of transparency and fairness, achieving explainability for meta-learning is crucial.  ...  Meta-learning is used to efficiently enable the automatic selection of machine learning models by combining data and prior knowledge.  ...  To integrate causality for a more precise interpretation, the latent factors for each feature in the causal structure model need to be discovered using the latent factors search.  ... 
arXiv:2205.10362v2 fatcat:2mo4kyd3onap5av4qyugz4f2ru

Biophysically motivated regulatory network inference: progress and prospects [article]

Richard Bonneau, Tarmo Aijo
2016 bioRxiv   pre-print
This perspective will focus on enumerating the elements of computational strategies that, when coupled to appropriate experimental designs, can lead to accurate large-scale models of chromatin-state and  ...  priors and data integration to constrain individual model parameters, estimation of latent regulatory factor activity under varying cell conditions, and new methods for learning and modeling regulatory  ...  These technologies can be combined to form multi-view data sets, that optimally combine perturbation, genetics and time series experimental designs into an overarching design.  ... 
doi:10.1101/051847 fatcat:3ril4ploovfbrkg7bj4pixtyt4

Allosteric Regulation at the Crossroads of New Technologies: Multiscale Modeling, Networks, and Machine Learning

Gennady M. Verkhivker, Steve Agajanian, Guang Hu, Peng Tao
2020 Frontiers in Molecular Biosciences  
The fundamental biological importance and complexity of these processes require a multi-faceted platform of synergistically integrated approaches for prediction and characterization of allosteric functional  ...  The wealth of structural and functional information along with diversity and complexity of allosteric mechanisms in therapeutically important protein families have provided a well-suited platform for development  ...  structure-based design of allosteric Hsp90 inhibitors.  ... 
doi:10.3389/fmolb.2020.00136 pmid:32733918 pmcid:PMC7363947 fatcat:vxoqxun6ebhdveqlbwi7l7rfui

Proactive Pseudo-Intervention: Causally Informed Contrastive Learning For Interpretable Vision Models [article]

Dong Wang, Yuewei Yang, Chenyang Tao, Zhe Gan, Liqun Chen, Fanjie Kong, Ricardo Henao, Lawrence Carin
2021 arXiv   pre-print
causal relevance.  ...  Further, our causally trained saliency maps are more succinct and meaningful relative to their non-causal counterparts.  ...  Machines instead learn from static observations that are unable to inform the structural dependencies for causal decisions.  ... 
arXiv:2012.03369v2 fatcat:bjioui6mnfhsjaudiny5vhcxg4

Biophysically Motivated Regulatory Network Inference: Progress and Prospects

Tarmo Äijö, Richard Bonneau
2016 Human Heredity  
or single-pathway experimental designs.  ...  Large-scale network inference approaches also have several advantages with regard to experimental design: that is, we can learn things from large-scale experimental designs that we cannot from single-gene  ...  These technologies can be combined to form multi-view data sets, that optimally combine perturbation, genetics and time series experimental designs into an overarching design.  ... 
doi:10.1159/000446614 pmid:28076866 fatcat:fytq7wm6dnbwjohtlg3bs5dbr4
« Previous Showing results 1 — 15 out of 5,751 results