Learning Stochastic Behaviour from Aggregate Data

Authors: Shaojun Ma, Shu Liu, Hongyuan Zha, Haomin Zhou

ICML 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate our approach in the context of a series of synthetic and real-world data sets.In this section, we evaluate our model on various synthetic and realistic data sets by employing Algorithm 1.
Researcher Affiliation Academia 1Department of Mathematics, Georgia Institute of Technology, Atlanta, GA 30332, USA 2School of Data Science, Shenzhen Research Institute of Big Data, The Chinese University of Hong Kong, Shenzhen, China
Pseudocode Yes Algorithm 1 Fokker Planck Process Algorithm
Open Source Code No No explicit statement or link providing access to the open-source code for the described methodology was found.
Open Datasets Yes Single-cell RNA-seq(Klein et al., 2015)
Dataset Splits No No explicit mention of a distinct validation set split. The paper states: 'We set the data size at each time point is N = 2000, treat 1200 data points as the training set and the other 800 data points as the test set' for synthetic data, and refers to 'training data' for realistic data without specifying a separate validation split.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, memory, or cloud instance types) used for running experiments were provided.
Software Dependencies No No specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions) were explicitly stated. It mentions 'Adam optimizer' but not its specific library version.
Experiment Setup Yes In all synthetic data experiments, we set the drift term g and the discriminator f as two simple fully-connected networks. The g network has one hidden layer and the f network has three hidden layers. Each layer has 32 nodes for both g and f. The only one activation function we choose is Tanh. In terms of training process, we use the Adam optimizer (Kingma & Ba, 2014) with learning rate 10 4. Furthermore, we use spectral normalization to realize f 1(Miyato et al., 2018). We initialize the weights with Xavier initialization(Glorot & Bengio, 2010) and train our model by Algorithm 1. We set the data size at each time point is N = 2000, treat 1200 data points as the training set and the other 800 data points as the test set, t is set to be 0.01.