No article text was recorded for this version.
unconstrained feature assumption
Version 1 of 134 papers40 conceptsfrozen 2026-10-04
Papers in this version
- Blog: On the Optimization Landscape of Neural Collapse under MSE Loss: Global Optimality with Unconstrained Features
- Blog: On the Optimization Landscape of Neural Collapse under MSE Loss: Global Optimality with Unconstrained Features
- Blog: Neural Collapse is Globally Optimal in Deep Regularized ResNets and Transformers
- Blog: Neural Collapse Under MSE Loss: Proximity to and Dynamics on the Central Path
- Blog: An Unconstrained Layer-Peeled Perspective on Neural Collapse
- Blog: A Message Passing Perspective on Learning Dynamics of Contrastive Learning
- Blog: Are All Losses Created Equal: A Neural Collapse Perspective
- Blog: On the emergence of simplex symmetry in the final and penultimate layers of neural network classifiers
- Blog: Neural Collapse with Normalized Features: A Geometric Analysis over the Riemannian Manifold
- Blog: Wide Neural Networks Trained with Weight Decay Provably Exhibit Neural Collapse
- Blog: Neural (Tangent Kernel) Collapse
- Blog: Perturbation Analysis of Neural Collapse
- Blog: Extended Unconstrained Features Model for Exploring Deep Neural Collapse
- Blog: Implicit Geometry of Next-token Prediction: From Language Sparsity Patterns to Model Representations
- Blog: Generalization bounds for deep convolutional neural networks
- Blog: Extended Unconstrained Features Model for Exploring Deep Neural Collapse
- Blog: Imbalance Trouble: Revisiting Neural-Collapse Geometry
- Blog: Stay-Positive: A Case for Ignoring Real Image Features in Fake Image Detection
- Blog: The Implicit Bias of Depth: From Neural Collapse to Softmax Codes
- Blog: Neural Collapse for Cross-entropy Class-Imbalanced Learning with Unconstrained ReLU Features Model
- Blog: Deep Neural Collapse Is Provably Optimal for the Deep Unconstrained Features Model
- Blog: A Global Geometric Analysis of Maximal Coding Rate Reduction
- Blog: Supervised Contrastive Representation Learning: Landscape Analysis with Unconstrained Features
- Blog: Neural Collapse under Gradient Flow on Shallow ReLU Networks for Orthogonally Separable Data
- Blog: A solvable model of learning generative diffusion: theory and insights
- Blog: Neural Collapse in Deep Linear Networks: From Balanced to Imbalanced Data
- Blog: From Linear to Nonlinear: Provable Weak-to-Strong Generalization through Feature Learning
- Blog: Landscape Connectivity and Dropout Stability of SGD Solutions for Over-parameterized Neural Networks
- Blog: Cross Entropy versus Label Smoothing: A Neural Collapse Perspective
- Blog: Can Kernel Methods Explain How the Data Affects Neural Collapse?
- Blog: A theoretical study of inductive biases in contrastive learning
- Blog: A Geometric Analysis of Neural Collapse with Unconstrained Features
- Blog: Average gradient outer product as a mechanism for deep neural collapse
- Blog: Unintentional Unalignment: Likelihood Displacement in Direct Preference Optimization
History & citation
Current · Permanent link
- This direction split from Geometric Simplification of Features in Deep Neural Networks (91% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_unconstrained_feature_assumption_6ec1453c,
title = {unconstrained feature assumption},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/unconstrained-feature-assumption/rdr_6ec1453ced424d29ac065dacf532e62c},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}This exact version (version 1, frozen text):
@misc{lacuna_unconstrained_feature_assumption_6ec1453c_v1,
title = {unconstrained feature assumption},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/unconstrained-feature-assumption/rdr_6ec1453ced424d29ac065dacf532e62c/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}