Convergence Guarantees for the DeepWalk Embedding on Block Models

Authors: Christopher Harker, Aditya Bhaskara

ICML 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental On the experimental side, we validate our results: we show a clear separation between the embeddings of vertices across clusters for different choices of the embedding dimension.
Researcher Affiliation Academia 1Kahlert School of Computing, University of Utah, Salt Lake City, UT, USA.
Pseudocode Yes Algorithm 1 Deep Walk Gradient Descent
Open Source Code No The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets No The paper uses graphs drawn from a "stochastic block model (SBM)" which is a generative model for graphs. It does not provide access information (link, DOI, specific citation) for a publicly available, pre-existing dataset used for training or evaluation.
Dataset Splits No The paper describes generating graphs from a stochastic block model for experiments. It does not mention explicit training, validation, or test splits of a fixed dataset, as the data is generated rather than partitioned from an existing source.
Hardware Specification No The paper does not provide any specific details regarding the hardware (e.g., CPU, GPU models, memory) used for running its experiments.
Software Dependencies No The paper does not provide specific version numbers for any software libraries, frameworks, or environments used in the experiments.
Experiment Setup Yes We run the algorithm for T = 100 iterations and used a learning rate of η = 0.01. The embeddings are initialized randomly so that x(t) 0.01 and y(t) 0.01. ... The learning rate was set for η = 1 n and training was rate for T = 75 iterations.