Expectation Propagation with Stochastic Kinetic Model in Complex Interaction Systems
Authors: Le Fang, Fan Yang, Wen Dong, Tong Guan, Chunming Qiao
NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we evaluate and benchmark the performance of our proposed algorithms (Algorithm 1) against mainstream state-of-the-art approaches. We have the flexibility to specify species, states, and events with different granularities in SKM, at either macroscopic or microscopic level. |
| Researcher Affiliation | Academia | Le Fang, Fan Yang, Wen Dong, Tong Guan, and Chunming Qiao Department of Computer Science and Engineering University at Buffalo {lefang, fyang24, wendong, tongguan, qiao}@buffalo.edu |
| Pseudocode | Yes | Algorithm 1 Make inference of a stochastic kinetic model with expectation propagation. Input: Discrete time SKM model (Eqs. (1),(2),(3)); Observation probabilities P(yt|xt) and initial values of αt, γt, λt for all populations m and time t. Expectation Propagation fixed point: Alternate between forward and backward iterations until convergence. |
| Open Source Code | No | Source code and a general function interface for other domains at both levels are here online |
| Open Datasets | No | We implement and benchmark algorithms on two representative datasets. In the Synth Town dataset, we synthesize a mini road network (Fig. 1(a)). Virtual residents go to work in the morning and back home in the evening. We synthesize their itineraries from MATSIM, a common Multi-agent transportation simulator[2]. In the Berlin dataset, we have a larger real world road network with 1,539 locations derived from Open Street Map and 9,178 people s itineraries synthesized from MATSIM. The paper mentions synthesizing data from MATSIM but does not provide a direct link or concrete access to the specific generated datasets used for the experiments. |
| Dataset Splits | No | The paper mentions training phases for neural networks but does not explicitly provide specific train/validation/test dataset splits (percentages, counts, or explicit standard split references) for reproduction. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments. |
| Software Dependencies | No | The paper mentions using MATSIM for data synthesis but does not provide specific version numbers for MATSIM or any other key software dependencies required to replicate the experiment. |
| Experiment Setup | No | The paper mentions tuning meta-parameters and empirically selecting the number of particles but does not provide specific hyperparameter values (e.g., learning rate, batch size, epochs, optimizer settings) or other detailed training configurations for reproducibility. |