Multiway Multislice PHATE: Visualizing Hidden Dynamics of RNNs through Training
Recurrent neural networks (RNNs) are a widely used tool for sequential data analysis, however, they are still often seen as black boxes of computation. Understanding the functional principles of these networks is critical to developing ideal model architectures and optimization strategies. Previous studies typically only emphasize the network representation post-training, overlooking their evolution process throughout training. Here, we present Multiway Multislice PHATE (MM-PHATE), a novel method for visualizing the evolution of RNNs' hidden states. MM-PHATE is a graph-based embedding using structured kernels across the multiple dimensions spanned by RNNs: time, training epoch, and units. We demonstrate on various datasets that MM-PHATE uniquely preserves hidden representation community structure among units and identifies information processing and compression phases during training. The embedding allows users to look under the hood of RNNs across training and provides an intuitive and comprehensive strategy to understanding the network's internal dynamics and draw conclusions, e.g., on why and how one model outperforms another or how a specific architecture might impact an RNN's learning ability.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Visualizing the PHATE of Neural Networks
Understanding why and how certain neural networks outperform others is key to guiding future development of network architectures and optimization methods. To this end, we introduce a novel visualization algorithm that r…
Continual LearningDimensionality ReductionMolecular dynamics study of the competitive binding of hydrogen peroxide and water molecules with the DNA phosphate groups
The hydrogen peroxide is present in the living cell at small concentrations that increase under the action of the heavy ion beams in the process of anticancer therapy. The interactions of hydrogen peroxide with DNA, prot…
Exploring epithelial-cell calcium signaling with geometric and topological data analysis
Spatio-temporal calcium imaging is widely used to study neuronal activity. However the function of calcium signaling in epithelial cells is not well understood. We thus explore the underlying fluorescence patterns using …
Topological Data AnalysisExploring the Geometry and Topology of Neural Network Loss Landscapes
Recent work has established clear links between the generalization performance of trained neural networks and the geometry of their loss landscape near the local minima to which they converge. This suggests that qualitat…
Dimensionality ReductionPhysics-Informed Neural Networks for Prognostics and Health Management of Lithium-Ion Batteries
For Prognostics and Health Management (PHM) of Lithium-ion (Li-ion) batteries, many models have been established to characterize their degradation process. The existing empirical or physical models can reveal important i…
Management