LAMVI-2: A Visual Tool for Comparing and Tuning Word Embedding Models

TL;DR

LAMVI-2 is a visual analytics tool designed to help developers compare and tune word embedding models more efficiently by visualizing high-level statistics and internal behaviors, addressing the complexity of hyperparameter tuning in deep learning.

Contribution

The paper introduces LAMVI-2, a novel visual system that facilitates comparison and tuning of word embedding models, integrating multiple views for comprehensive analysis.

Findings

LAMVI-2 enables faster model comparison and selection.

Developers can better understand internal model behaviors.

The tool improves accuracy in hyperparameter tuning.

Abstract

Tuning machine learning models, particularly deep learning architectures, is a complex process. Automated hyperparameter tuning algorithms often depend on specific optimization metrics. However, in many situations, a developer trades one metric against another: accuracy versus overfitting, precision versus recall, smaller models and accuracy, etc. With deep learning, not only are the model's representations opaque, the model's behavior when parameters "knobs" are changed may also be unpredictable. Thus, picking the "best" model often requires time-consuming model comparison. In this work, we introduce LAMVI-2, a visual analytics system to support a developer in comparing hyperparameter settings and outcomes. By focusing on word-embedding models ("deep learning for text") we integrate views to compare both high-level statistics as well as internal model behaviors (e.g., comparing word 'distances'). We demonstrate how developers can work with LAMVI-2 to more quickly and accurately narrow down an appropriate and effective application-specific model.