The Data Manifold under the Microscope
Paper • 2606.15760 • Published • 2 •
5
Marios Koulakis
koulakis
·
AI & ML interests
I work on deep learning theory, data and representation geometry, manifold learning, dimensionality reduction, clustering, mechanistic interpretability, and trustworthy AI, with a focus on geometric benchmarks for learned representations.
Recent Activity
commentedon a paper about 3 hours ago
The Data Manifold under the Microscope submitted a paper 1 day ago
The Data Manifold under the Microscope authored a paper 1 day ago
Domain-independent detection of known anomaliesOrganizations
None yet
commented a
paper about 3 hours ago
submitted a
paper to Daily Papers 1 day ago
authored 6
papers 1 day ago
Domain-independent detection of known anomalies
Paper • 2407.02910 • Published
Hierarchical Nearest Neighbor Graph Embedding for Efficient Dimensionality Reduction
Paper • 2203.12997 • Published
Position: Quo Vadis, Unsupervised Time Series Anomaly Detection?
Paper • 2405.02678 • Published
The Data Manifold under the Microscope
Paper • 2606.15760 • Published • 2
Is my Driver Observation Model Overconfident? Input-guided Calibration Networks for Reliable and Interpretable Confidence Estimates
Paper • 2204.04674 • Published
Affect-DML: Context-Aware One-Shot Recognition of Human Affect using Deep Metric Learning
Paper • 2111.15271 • Published