Learning with Temporal Point Processes
ICML 2018 Tutorial
Basic details
In recent years, there has been an increasing number of machine learning models, inference methods and control algorithms using temporal point processes (TPPs). They have been particularly popular for understanding, predicting, and enhancing the functioning of social and information systems, where they have achieved unprecedented performance. This tutorial aims to introduce temporal point processes to the machine learning community at large.
In the first part of the tutorial, we will first provide an introduction to the basic theory of temporal point processes, then revisit several types of points processes, and finally introduce advanced concepts such as marks and dynamical systems with jumps. In the second and third parts of the tutorial, we will explain how temporal point processes have been used in developing a variety of recent machine learning models and control algorithms, respectively. Therein, we will revisit recent advances related to, e.g., deep learning, Bayesian nonparametrics, causality, stochastic optimal control and reinforcement learning. In each of the above parts, we will highlight open problems and future research to facilitate further research in temporal point processes within the machine learning community.
Timeline
Resources
- Uncovering the Temporal Dynamics of Diffusion Networks (Models/Inference)
- Modeling Information Propagation with Survival Theory (Models/Inference)
- Estimating Diffusion Network Structures: Recovery Conditions, Sample Complexity & Soft-thresholding Algorithm (Models/Inference)
- Coevolve: A Joint Point Process Model for Information Diffusion and Network Co-evolution (Models/Inference)
- Recurrent Marked Temporal Point Process: Embedding Event History to Vector (Models/Inference)
- Learning and Forecasting Opinion Dynamics in Social Networks (Models/Inference)
- Uncovering the Temporal Dynamics of Diffusion Networks (Models/Inference)
- A dirichlet mixture model of hawkes processes for event sequence clustering (Models/Inference)
- Dirichlet-hawkes processes with applications to clustering continuous-time document streams (Models/Inference)
- Modeling the Dynamics of Learning Activity (Models/Inference)
- Neural survival recommender (Models/Inference)
- Recurrent coevolutionary feature embedding processes for recommendation (Models/Inference)
- The neural hawkes process: A neurally self-modulating multivariate point process (Models/Inference)
- Know-evolve: Deep temporal reasoning for dynamic knowledge graphs (Models/Inference)
- Learning conditional generative models for temporal point processes (Models/Inference)
- Wasserstein learning of deep generative point process models (Models/Inference)
- Modeling the Dynamics of Learning Activity (Models/Inference)
- Uncovering causality from multivariate hawkes integrated cumulants (Models/Inference)
- On the causal effect of badges (Models/Inference)
- Distilling Learning granger causality for hawkes processes (Models/Inference)
- Smart Broadcasting: Do You Want to be Seen? (RL/Control)
- RedQueen: An Online Algorithm for Smart Broadcasting in Social Networks (RL/Control)
- Cheshire: An Online Algorithm for Activity Maximization in Social Networks (RL/Control)
- Steering Social Activity: A Stochastic Optimal Control Point of View (RL/Control)
- Leveraging the Crowd to Detect and Reduce the Spread of Fake News and Misinformation (RL/Control)
- Optimizing Human Learning (RL/Control)
- Variational Policy for Guiding Point Processes (RL/Control)
- A Stochastic Differential Equation Framework for Guiding Online User Activities in Closed Loop (RL/Control)
- Deep Reinforcement Learning of Marked Temporal Point Processes (RL/Control)
- Fake News Mitigation via Point Process Based Intervention (RL/Control)
- PtPack (C++ TPP library)
- Tick (Python TPP library)
- NL MPI-SWS (Several TPP models/control algorithms)
