Deeper Connections between Neural Networks and Gaussian Processes Speed-up Active Learning
Authors: Evgenii Tsymbalov, Sergei Makarychev, Alexander Shapeev, Maxim Panov
IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experimental study is focused on real-world data to ensure that the algorithms are indeed useful in real-world scenarios. We compare the following algorithms: pure random sampling; sampling based on the variance of NN stochastic output from ˆf, which we refer to as MCDUE (see [Gal et al., 2017; Tsymbalov et al., 2018]); the proposed GP-based approaches (NNGP). |
| Researcher Affiliation | Academia | Evgenii Tsymbalov , Sergei Makarychev , Alexander Shapeev and Maxim Panov Skolkovo Institute of Science and Technology (Skoltech) {e.tsymbalov, sergei.makarychev, a.shapeev, m.panov}@skoltech.ru |
| Pseudocode | No | The paper describes the NNGP active learning procedure, and includes Figure 3 as a 'Schematic representation of the NNGP approach to active learning', but it does not provide any formal pseudocode or a clearly labeled algorithm block. |
| Open Source Code | No | The paper mentions using a 'TensorFlow implementation of a Sch Net' which implies using an existing framework, but it does not provide any statement or link for the open-sourcing of the authors' own methodology or implementation code. |
| Open Datasets | Yes | Following this paper, we use the airline delays dataset (see [Hensman et al., 2013]) and we conducted a series of experiments with active learning performed on the data from the UCI ML repository 1. 1 D. Dua, E. Taniskidou. UCI Machine Learning Repository [http://archive.ics.uci.edu/ml] and We tested our approach on the problem of predicting the internal energy of the molecule at 0K from the QM9 data set [Ramakrishnan et al., 2014]. |
| Dataset Splits | Yes | For every experiment, data are shuffled and split in the following proportions: 10% for the training set Dtrain, 5% for the test set Dtest, 5% for the validation set Dval needed for early-stopping and 80% for the pool P. |
| Hardware Specification | No | The paper describes the neural network architectures and training procedures, but it does not provide any specific details about the hardware (e.g., GPU models, CPU types, or memory) used for running the experiments. |
| Software Dependencies | No | The paper mentions using a 'TensorFlow implementation' for Sch Net, but it does not specify the version number for TensorFlow or any other software dependencies crucial for replication. |
| Experiment Setup | Yes | NN consisting of two layers with 50 neurons each, leaky Re LU activation function, and trained with respect to NCP-based loss function. (Airline Delays Dataset) and We used a simple neural network with three hidden layers of sizes 256, 128 and 128. (UCI Datasets) and On each active learning iteration, we perform 100 000 training epochs (Sch Net Training). |