Filters








173,854 Hits in 4.8 sec

Intensity-Free Learning of Temporal Point Processes [article]

Oleksandr Shchur, Marin Biloš, Stephan Günnemann
2020 arXiv   pre-print
Temporal point processes are the dominant paradigm for modeling sequences of events happening at irregular intervals.  ...  The standard way of learning in such models is by estimating the conditional intensity function. However, parameterizing the intensity function usually incurs several trade-offs.  ...  Learning temporal point processes.  ... 
arXiv:1909.12127v2 fatcat:6fgz2rfrc5fabd3sszdyub6fiq

Understanding the Spread of COVID-19 Epidemic: A Spatio-Temporal Point Process View [article]

Shuang Li, Lu Wang, Xinyun Chen, Yixiang Fang, Yan Song
2021 arXiv   pre-print
In this paper, we model the propagation of the COVID-19 as spatio-temporal point processes and propose a generative and intensity-free model to track the spread of the disease.  ...  In comparison with the traditional likelihood-based learning methods, this imitation learning framework does not need to prespecify an intensity function, which alleviates the model-misspecification.  ...  Generate New Events Our generative spatio-temporal point process model is intensity-free.  ... 
arXiv:2106.13097v1 fatcat:jkfrjbolxnfnne5yxnelw2eil4

Wasserstein Learning of Deep Generative Point Process Models [article]

Shuai Xiao, Mehrdad Farajtabar, Xiaojing Ye, Junchi Yan, Le Song, Hongyuan Zha
2017 arXiv   pre-print
In this paper, we propose an intensity-free approach for point processes modeling that transforms nuisance processes to a target one.  ...  Furthermore, we train the model using a likelihood-free leveraging Wasserstein distance between point processes.  ...  Temporal Point Processes A particularly interesting case of point processes is given when S is the time interval [0, T ), which we will call a temporal point process.  ... 
arXiv:1705.08051v1 fatcat:fdon3bysw5d63cyirpowhoglu4

Point Process Flows [article]

Nazanin Mehrasa, Ruizhi Deng, Mohamed Osama Ahmed, Bo Chang, Jiawei He, Thibaut Durand, Marcus Brubaker, Greg Mori
2019 arXiv   pre-print
We propose an intensity-free framework that directly models the point process distribution by utilizing normalizing flows.  ...  Event sequences can be modeled by temporal point processes (TPPs) to capture their asynchronous and probabilistic nature.  ...  an intensity-free flow framework to model the timing of events in point process sequences.  ... 
arXiv:1910.08281v3 fatcat:ihvpycyk4jcjjce2ruuu7ojj5u

Data-driven chimney fire risk prediction using machine learning and point process tools [article]

C. Lu, M.N.M. van Lieshout, M. de Graaf, P. Visscher
2021 arXiv   pre-print
In this paper, we develop a combined machine learning and statistical modeling process to predict chimney fires.  ...  Secondly, we design a Poisson point process model and apply associated logistic regression estimation to estimate the parameters.  ...  We also thank Emiel Borggreve, Niels Peters and Etienne Mulder from the Twente Fire Brigade for their help with data cleaning and pre-processing.  ... 
arXiv:2112.07257v1 fatcat:47koqaciqrbxxiobaso2vj6u64

ChOracle: A Unified Statistical Framework for Churn Prediction [article]

Ali Khodadadi, Seyed Abbas Hosseini, Ehsan Pajouheshgar, Farnam Mansouri, Hamid R. Rabiee
2019 arXiv   pre-print
In this paper, we introduce ChOracle, an oracle that predicts the user churn by modeling the user return times to service by utilizing a combination of Temporal Point Processes and Recurrent Neural Networks  ...  We also develop an efficient approximate variational algorithm for learning parameters of the proposed RNN by using back propagation through time.  ...  It learns a vector of length 4320 for each event that is the intensity function of a temporal point process for about 180 days.  ... 
arXiv:1909.06868v1 fatcat:v6wuzit5qnfq7b63kl2tw6sdbi

Learning the statistics of pain: computational and neural mechanisms [article]

Flavia Mancini, Suyi Zhang, Ben Seymour
2021 bioRxiv   pre-print
In conclusion, this study extends what is conventionally considered a sensory pain pathway dedicated to process pain intensity, to include the generation of Bayesian internal models of temporal statistics  ...  We demonstrate that humans can learn to extract these regularities, and explicitly predict the likelihood of forthcoming pain intensities in a manner consistent with optimal Bayesian models with dynamic  ...  We are grateful to Professor Zoe Kourtzi and Dr Michael Lee for helpful discussions about the concept of the study, and to the staff of the Wolfson Brain Imaging Centre for their support during data collection  ... 
doi:10.1101/2021.10.21.465270 fatcat:me4o4vhe6fgdrk6azmm3cwfxva

Predicting ultrafast nonlinear dynamics in fibre optics with a recurrent neural network [article]

Lauri Salmela, Nikolaos Tsipinakis, Alessandro Foi, Cyril Billet, John M. Dudley, Goëry Genty
2020 arXiv   pre-print
Comparison with experiments for the case of soliton compression shows remarkable agreement in both temporal and spectral domains.  ...  , solely from a given transform-limited input pulse intensity profile.  ...  The simulations use 1024 spectral/temporal grid points with temporal window size of 10 ps and a step size of 0.13 mm (10,000 steps).  ... 
arXiv:2004.14126v1 fatcat:42plvgo6qffibl4f325bg7ohbq

Ghost cytometry

Sadao Ota, Ryoichi Horisaki, Yoko Kawamura, Masashi Ugawa, Issei Sato, Kazuki Hashimoto, Ryosuke Kamesawa, Kotaro Setoyama, Satoko Yamaguchi, Katsuhito Fujiu, Kayo Waki, Hiroyuki Noji
2018 Science  
Combinatorial use of the temporal waveform with the intensity distribution of the random pattern allows us to computationally reconstruct cell morphology.  ...  More importantly, we show that applying machine-learning methods directly on the compressed waveforms without image reconstruction enables efficient image-free morphology-based cytometry.  ...  extracted from the temporally modulated signals of fluorescence intensity.  ... 
doi:10.1126/science.aan0096 pmid:29903975 fatcat:qz2l7aak7baxlee5klryzevep4

Representation Learning over Dynamic Graphs [article]

Rakshit Trivedi, Mehrdad Farajtabar, Prasenjeet Biswal, Hongyuan Zha
2018 arXiv   pre-print
We employ a time-scale dependent multivariate point process model to capture these dynamics.  ...  The learned embeddings drive the dynamics of two key processes namely, communication and association between nodes in dynamic graphs.  ...  Following that work, there have been increased attraction in topic of learning conditional intensity function using deep learning [28] and also intensity free approach using GANS [46] for learning  ... 
arXiv:1803.04051v2 fatcat:gcgfprjh6fhpbhgcegbou4aiee

Randomized Controlled Trial Considering Varied Exercises for Reducing Proactive Memory Interference

Emily Frith, Eveleen Sng, Paul Loprinzi
2018 Journal of Clinical Medicine  
We used the Rey Auditory Verbal Learning Test (RAVLT) to assess proactive memory interference.  ...  In this sample of young adults, high intensity exercise prior to memory encoding showed a non-significant tendency to attenuate impairments in recall attributable to proactive memory interference.  ...  The mPFC differentiates memory cues at the point of encoding, which assists correct retrieval [5] .  ... 
doi:10.3390/jcm7060147 pmid:29891765 pmcid:PMC6024907 fatcat:pjamfc5rqba5pklhb6zhndqbqm

DyRep: Learning Representations over Dynamic Graphs

Rakshit Trivedi, Mehrdad Farajtabar, Prasenjeet Biswal, Hongyuan Zha
2019 International Conference on Learning Representations  
Concretely, we propose a two-time scale deep temporal point process model that captures the interleaved dynamics of the observed processes.  ...  We present DyRep -a novel modeling framework for dynamic graphs that posits representation learning as a latent mediation process bridging two observed processes namely -dynamics of the network (realized  ...  This work was supported in part by NSF IIS-1717916, NSF CMMI-1745382, NSFC-Zhejiang Joint Fund for the Integration of Industrialization and Information (U1609220) and National Science Foundation of China  ... 
dblp:conf/iclr/TrivediFBZ19 fatcat:z6og4ca52rawpmj2bnubg7vdfq

Machine learning for prediction of extreme statistics in modulation instability [article]

Mikko Närhi, Lauri Salmela, Juha Toivonen, Cyril Billet, John M. Dudley, Goëry Genty
2018 arXiv   pre-print
Here, we show how Machine Learning can overcome this limitation by predicting statistics for the maximum intensity of temporal peaks in modulation instability based only on spectral measurements.  ...  Specifically, we train a neural network based Machine Learning model to correlate spectral and temporal properties of optical fibre modulation instability using data from numerical simulations, and we  ...  In the context of relating MI dynamics to the appearance of extreme events and rogue waves, our aim is to predict the intensity of the maximum peak occurring in a given temporal profile (i.e. the points  ... 
arXiv:1806.06121v1 fatcat:a4ho6yzeovcyvnjzgbi62tguya

Intrinsic Dimensionality Predicts the Saliency of Natural Dynamic Scenes

E. Vig, M. Dorr, T. Martinetz, E. Barth
2012 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Using a large collection of eye movements on high-resolution videos, supervised learning techniques fine-tune the free parameters whose addition is inevitable with increasing complexity.  ...  Index Terms-Computational models of vision, video analysis, computer vision, spatio-temporal saliency, eye movement prediction, intrinsic dimension, visual attention, interest point detection.  ...  IST-C-033816, see www.gazecom.eu) of the 6th Framework Programme. All views expressed herein are those of the authors alone; the European Community is not liable for any use made of the information.  ... 
doi:10.1109/tpami.2011.198 pmid:22516647 fatcat:wb46op4h45d4thpjpaunkwppvq

Machine Learning Models for Classifying Physical Activity in Free-Living Preschool Children

Matthew N. Ahmadi, Toby G. Pavey, Stewart G. Trost
2020 Sensors  
The inclusion of temporal features increased the accuracy, mainly for the wrist classification models, by an average of 6.2 percentage points.  ...  Training new models on free-living accelerometer data, reducing the number of prediction windows comprised of multiple activity types by using shorter windows, including temporal features such as standard  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/s20164364 pmid:32764316 fatcat:3ngrl5u3yvanpcbts23ahybqey
« Previous Showing results 1 — 15 out of 173,854 results