Uncertainty Quantification for Inferring Hawkes Networks
Authors: Haoyun Wang, Liyan Xie, Alex Cuozzo, Simon Mak, Yao Xie
NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we present a numerical example based on synthetic data to demonstrate the performance of the proposed confidence intervals. We compare the coverage ratio of the confidence intervals: the percentage of confidence intervals that contain the true parameters, for the same nominal confidence level (1 ε). |
| Researcher Affiliation | Academia | 1Georgia Institute of Technology, 2 Duke University |
| Pseudocode | Yes | Algorithm 1 summarizes how to find the concentration-bound based confidence set. Algorithm 1: Polyhedral Confidence Set for αi |
| Open Source Code | No | The paper does not provide concrete access to source code (specific repository link, explicit code release statement, or code in supplementary materials) for the methodology described in this paper. |
| Open Datasets | No | The neural spike train data is simulated via the Py NN Python package [5] with the NEURON simulator [4], which was chosen over in vivo recordings for straightforward data collection. The paper describes data generation rather than the use of an existing public or open dataset with access information. |
| Dataset Splits | No | The paper describes simulating data for constructing CIs but does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce data partitioning. |
| Hardware Specification | No | The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper mentions 'Py NN Python package [5]' and 'NEURON simulator [4]' as tools used for data simulation, but it does not provide specific version numbers for these or other ancillary software components. |
| Experiment Setup | Yes | The experimental set-up is as follows. The neural spike train data is simulated via the Py NN Python package [5] with the NEURON simulator [4]... The neuronal network consists of excitatory and inhibitory networks in a ratio of 4 to 1, which are connected sparsely and at random. The neurons are modeled as exponential integrate-and-fire neurons with default parameters... Using the above network structure with D = 32 neurons, we simulate a long sequence (2000 seconds) of spiking data, and fit a Hawkes network using an exponential influence function with a decay rate of 1 millisecond. |