Meta Learning with Relational Information for Short Sequences
Authors: Yujia Xie, Haoming Jiang, Feng Liu, Tuo Zhao, Hongyuan Zha
NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Numerical experiments on both synthetic and real data show that HARMLESS outperforms existing methods in terms of predicting the future events. |
| Researcher Affiliation | Academia | Yujia Xie College of Computing, Georgia Tech Xie.Yujia000@gmail.com; Haoming Jiang College of Engineering, Georgia Tech jianghm@gatech.edu; Feng Liu Florida Atlantic University FLIU2016@fau.edu; Tuo Zhao College of Engineering, Georgia Tech tuo.zhao@isye.gatech.edu; Hongyuan Zha0 Institute for Data and Decision Analytics, the Chinese University of Hong Kong, Shenzhen Shenzhen Institute of Artificial Intelligence and Robotics for Society zhahy@cuhk.edu.cn |
| Pseudocode | No | The paper describes updates and an algorithm but does not contain a structured pseudocode or algorithm block (clearly labeled algorithm sections or code-like formatted procedures) in the main text. |
| Open Source Code | No | The paper does not contain any explicit statement about releasing source code or provide a link to a code repository for the methodology described. |
| Open Datasets | Yes | Linked In dataset (Xu et al., 2017b) contains job hopping records of the users. Math Overflow dataset (Paranjape et al., 2017) contains records of the users posting and answering math questions. |
| Dataset Splits | Yes | We hold out the last timestamp of each sequence, and split the hold-out timestamps into a validation set and a test set. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details, such as library or solver names with version numbers, needed to replicate the experiment. |
| Experiment Setup | Yes | Following Rasmussen (2013); Zhou et al. (2013), we choose exponential impact function g(t; {δ, !}) = δ!e !t. The conditional intensity function is λ(t; , ) = µ + δ!e !(t (m)). Note that each Hawkes process model only contains three parameters, µ, δ, and !. To keep the parameters non-negative, in practice we replace log Li(e (i)k ) with a regularized loglikelihood in update (10), log Li(e (i)k ) , log Li(e (i)k ) + R(e (i)k ) , log Li(e (i)k ) + log(e (i)k ) + log(e!(i)k ). |