6 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…
Scalable and Interpretable Representation Alignment with Ordinal Similarity
Diogo Soares, Pankhil Gawade, Andrea Dittadi +1
Evaluating representation similarity is fundamental to representation learning. However, existing metrics suffer from significant limitations: they lack interpretability due to shi…
Sample Efficient Generative Molecular Optimization with Joint Self-Improvement
Serra Korkmaz, Adam Izdebski, Jonathan Pirnay +5
Generative molecular optimization aims to design molecules with properties surpassing those of existing compounds. However, such candidates are rare and expensive to evaluate, yiel…
seqme: a Python library for evaluating biological sequence design
Rasmus Møller-Larsen, Adam Izdebski, Jan Olszewski +4
Recent advances in computational methods for designing biological sequences have sparked the development of metrics to evaluate these methods performance in terms of the fidelity o…
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…
Synergistic Benefits of Joint Molecule Generation and Property Prediction
Adam Izdebski, Jan Olszewski, Pankhil Gawade +5
Modeling the joint distribution of data samples and their properties allows to construct a single model for both data generation and property prediction, with synergistic benefits…