• Publications
  • Influence
More Robust Doubly Robust Off-policy Evaluation
TLDR
We study the problem of off-policy evaluation (OPE) in reinforcement learning, where the goal is to estimate the performance of a policy from the data generated by another policy(ies). Expand
Learning Granger Causality for Hawkes Processes
TLDR
We propose an effective learning algorithm combining a maximum likelihood estimator (MLE) with a sparse-group-lasso (SGL) regularizer. Expand
Dirichlet-Hawkes Processes with Applications to Clustering Continuous-Time Document Streams
TLDR
We propose a novel random process, referred to as the Dirichlet-Hawkes process, to take into account both information in a unified framework. Expand
Improved Knowledge Distillation via Teacher Assistant: Bridging the Gap Between Student and Teacher
TLDR
We introduce multistep knowledge distillation which employs an intermediate-sized network (a.k.a. teacher assistant) to bridge the gap between the student and teacher. Expand
Improved Knowledge Distillation via Teacher Assistant
TLDR
We show that the student network performance degrades when the gap between student and teacher is large. Expand
COEVOLVE: A Joint Point Process Model for Information Diffusion and Network Co-evolution
TLDR
In this work, we propose a probabilistic generative model, Coevolve, for the joint dynamics of information diffusion and network evolution. Expand
Shaping Social Activity by Incentivizing Users
TLDR
We model social events using multivariate Hawkes processes, which can capture both endogenous and exogenous event intensities, and derive a time dependent linear relation between the intensity of exogenous events and the overall network activity. Expand
Fake News Mitigation via Point Process Based Intervention
TLDR
We propose the first multistage intervention framework that tackles fake news in social networks by combining reinforcement learning with a point process network activity model. Expand
Wasserstein Learning of Deep Generative Point Process Models
TLDR
In this paper, we propose an intensity-free approach for point processes modeling that transforms nuisance processes to a target one. Expand
...
1
2
3
4
5
...