8 papers
A Closed Form Depth-Width Law for Neural Collapse in Deep Linear Networks
Pankhil Gawade, Kirsten Fischer, Guy Wolf +1
How depth and width of neural networks shape learned representations is usually studied only after training. We show that, in a tractable feature-learning setting, this dependence…
Forest-Guided Semantic Transport for Label-Supervised Manifold Alignment
Adrien Aumon, Myriam Lizotte, Guy Wolf +2
Label-supervised manifold alignment bridges the gap between unsupervised and correspondence-based paradigms by leveraging shared label information to align multimodal datasets. How…
Revisiting Forest Proximities via Sparse Leaf-Incidence Kernels
Adrien Aumon, Guy Wolf, Kevin R. Moon +1
Decision forests induce supervised similarities through the partition structure of their trees. Yet forest proximity computation is still often treated as a quadratic operation in…
Freeze, Diffuse, Decode: Task-Aware Adaptation of Transformer Embeddings for Antimicrobial Peptide Design
Pankhil Gawade, Adam Izdebski, Myriam Lizotte +4
Pretrained transformers provide rich, general-purpose embeddings, which are transferred to downstream tasks. However, current transfer strategies: fine-tuning and probing, either d…
Random Forest Autoencoders for Guided Representation Learning
Adrien Aumon, Shuang Ni, Myriam Lizotte +3
Extensive research has produced robust methods for unsupervised data visualization. Yet supervised visualization$\unicode{x2013}$where expert labels guide representations$\unicode{…
Enhancing Supervised Visualization through Autoencoder and Random Forest Proximities for Out-of-Sample Extension
Shuang Ni, Adrien Aumon, Guy Wolf +2
The value of supervised dimensionality reduction lies in its ability to uncover meaningful connections between data features and labels. Common dimensionality reduction methods emb…
Noisy Data Visualization using Functional Data Analysis
Haozhe Chen, Andres Felipe Duque Correa, Guy Wolf +1
Data visualization via dimensionality reduction is an important tool in exploratory data analysis. However, when the data are noisy, many existing methods fail to capture the under…
Spectral Temporal Contrastive Learning
Sacha Morin, Somjit Nath, Samira Ebrahimi Kahou +1
Learning useful data representations without requiring labels is a cornerstone of modern deep learning. Self-supervised learning methods, particularly contrastive learning (CL), ha…