Recurrent Marked Temporal Point Processes: Embedding Event History to Vector

By Nan Du, Hanjun Dai, Rakshit S. Trivedi, U. Upadhyay, Manuel Gomez-Rodriguez, Le Song. 2016

831 citations recorded by the source corpus.

Subfields: Applied Mathematics, Computational Mechanics, Geometry and Topology, Model Architectures, Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Applied Mathematics

In Computational Mechanics

In Geometry and Topology

In Model Architectures

In Probabilistic & Bayesian Methods

Citations in the corpus

Neural Ordinary Differential Equations (2018, neutral): “We trained a second version of this RNN whose inputs were concatenated with the time difference to the next observation to aid RNN with irregular observations.”
Learning from History: Modeling Temporal Knowledge Graphs with Sequential Copy-Generation Networks (2020, background): “transaction recommendation (Lortscher Jr 2010), event process induction (Bouchard and Andreoli 2012; Du et al. 2016) and social relation prediction (Zhou et al.”
Neural Survival Recommender (2017, background): “Finally, [7] proposed recurrent networks-based models to solve to model a sequence of events, but they did not attempt to solve the recommendation problems and return time prediction jointly.”
Disease-Atlas: Navigating Disease Trajectories using Deep Learning (2018, methodology): “This includes joint architectures for forecasting follow-up times and clinical diagnoses (Choi et al., 2016; Du et al., 2016) and multitask learning (Harutyunyan et al.”
Know-Evolve: Deep Temporal Reasoning for Dynamic Knowledge Graphs (2017, methodology): “Previous techniques (Du et al., 2016; Hidasi et al., 2016) that use BPTT algorithm decompose data into independent sequences and train on mini-batches of those sequences.”
Relational Representation Learning for Dynamic (Knowledge) Graphs: A Survey (2019, background): “The function usually contains learnable parameters [64] that can be learned from the data.”
Self-attention with Functional Time Representation Learning (2019, methodology): “Several methods are proposed to couple point process with RNNs to take account of temporal information [23, 22, 14, 6].”
DeepMove: Predicting Human Mobility with Attentional Recurrent Networks (2018, methodology): “[9] propose Recurrent Marked Temporal Point Process (RMTPP) to learn the conditional intensity function automatically from the history.”
DeepHawkes: Bridging the Gap between Prediction and Understanding of Information Cascades (2017, background): “to learn the function of arrival rate in point process [30, 31].”
Representation Learning for Dynamic Graphs: A Survey (2019, methodology): “Recently, there has been growing interest in learning the intensity function entirely from the data (see, e.g., Du et al. (2016)).”
Deep Learning for Time Series Forecasting: Tutorial and Literature Survey (2020, background): “Finally, a recent trend in sequence modelling employs NNs in modelling discrete event sequences observed in continuous time [48, 133, 164, 165, 182, 194] and [166] for an overview.”
Process (2020, methodology): “There are have been attempts on RNNs based point process models [1, 2, 5, 7], which use RNNs to memorize the influence of historical events.”
GRU-ODE-Bayes: Continuous modeling of sporadically-observed time series (2019, background): “To overcome this issue, the main approach consists in some form of data imputation and jointly feeding the observation mask and times of observations to the recurrent network (Che et al., 2018; Choi et al., 2016a; Lipton et al., 2016; Du et al., 2016; Choi et al., 2016b; Cao et al., 2018).”
Time2Vec: Learning a Vector Representation of Time (2019, methodology): “When time is known to be a relevant feature, it is often fed in as yet another input dimension [10, 14, 35].”
Evaluation of session-based recommendation algorithms (2018, methodology): “uence modeling approaches are usually based on Markov Chain (MC) models [14,18,23,36], reinforcement learning (RL) and Markov Decision Processes (MDP) [39, 46, 53], or Recurrent Neural Networks (RNN) [11, 21, 22, 35, 47, 49, 50, 55, 60, 63]. Again, the typical application scenarios of these methods ”
Sequence-Aware Recommender Systems (2018, methodology): “tem features with user temporal features into a joint model. Static features are modeled by several feed-forward neural networks, whereas the temporal features are modeled by a set of RNNs. Du et al. [29] address the list continuation problem in a wide range of application domains, from taxi drop-of”