On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

count % of corpus

7 results for “CSQ”

Faster Algorithms for Agnostically Learning Disjunctions and their Implicationscs.LG

New algorithm learns disjunctions faster than previous methods.

problem Learning Boolean disjunctions in the agnostic PAC model.

method Developed an agnostic learner with complexity

2^{ ilde{O}(n^{1/3})}

result First separation between SQ and CSQ models in distribution-free agnostic learning.

Boolean disjunctions agnostic learning polynomial regression statistical query model CSQ model

2025-04-21abs ↗pdf ↗

Agnostic Learning of Arbitrary ReLU Activation under Gaussian Marginalscs.LG

Algorithm learns arbitrary ReLU neurons under Gaussian inputs.

problem Learn an arbitrary ReLU activation over Gaussian marginals.

method Statistical Query (SQ) algorithm that outputs a ReLU activation achieving

O(\mathrm{OPT}) + \varepsilon

loss.

result First constant factor approximation for arbitrary bias in polynomial time.

ReLU Neural Networks Statistical Query CSQ Gaussian Marginals

2024-11-21abs ↗pdf ↗

Pruning is Optimal for Learning Sparse Features in High-Dimensionsstat.ML

Pruning neural networks improves feature learning in high-dimensional models.

problem Improving feature quality in high-dimensional statistical models.

method Demonstrated that pruning neural networks can optimally learn sparse models using gradient descent.

result Pruning neural networks proportional to the sparsity level of the model directions improves sample complexity.

pruning sparse features high-dimensional models gradient descent statistical models

2024-06-12abs ↗pdf ↗

SGD learning on neural networks: leap complexity and saddle-to-saddle dynamicscs.LG

SGD learns neural networks with a complexity measure called leap.

problem Time complexity of SGD learning on neural networks.

method Introduced a complexity measure called leap, proved conjecture for Gaussian data, and showed saddle-to-saddle dynamics.

result Proved a conjecture about the time complexity of learning functions with low-dimensional support.

SGD neural networks time complexity leap saddle-to-saddle dynamics

2023-02-21abs ↗pdf ↗

On the hardness of learning under symmetriescs.LG

Gradient descent struggles to learn equivariant neural networks, even with symmetries.

problem Learning equivariant neural networks via gradient descent is hard.

method Lower bounds for various equivariant neural network classes.

result Gradient descent struggles to learn equivariant neural networks, even with symmetries.

equivariant neural networks gradient descent hardness symmetry lower bounds

2024-01-03abs ↗pdf ↗

Smoothing the Landscape Boosts the Signal for SGD: Optimal Sample Complexity for Learning Single Index Modelscs.LG

Improved SGD learning for single index models reduces sample complexity.

problem Learning a single index model with optimal sample complexity.

method Using smoothed loss in online SGD to reduce sample complexity.

result Online SGD with smoothed loss achieves optimal sample complexity of

d^{k^\star/2}

SGD Single index model Optimal sample complexity Smoothed loss Hermite coefficients

2023-05-18abs ↗pdf ↗

Mamba Can Learn Low-Dimensional Targets In-Context via Test-Time Feature Learningcs.LG

Mamba efficiently learns low-dimensional targets in-context via feature extraction.

problem Learning low-dimensional targets in context for computational efficiency.

method Test-time feature learning of a single-index model using Mamba's pretrained linear-time sequence model.

result Mamba achieves efficient in-context learning of low-dimensional targets via feature extraction.

Mamba in-context learning feature extraction single-index model test-time feature learning

2025-10-14abs ↗pdf ↗