Social Welfare and Profit Maximization from Revealed Preferences von Ji, Ziwei, Mehta, Ruta, Telgarsky, Matus

Volltext

Feature selection with gradient descent on two-layer networks in low-rotation regimes von Telgarsky, Matus

Volltext bestellen

Stochastic linear optimization never overfits with quadratically-bounded losses on general data von Telgarsky, Matus

Volltext bestellen

On Achieving Optimal Adversarial Test Error von Li, Justin D, Telgarsky, Matus

Volltext bestellen

Feature selection with gradient descent on two-layer networks in low-rotation regimes von Matus Telgarsky

Volltext

One-layer transformers fail to solve the induction heads task von Sanford, Clayton, Hsu, Daniel, Telgarsky, Matus

Volltext bestellen

Stochastic linear optimization never overfits with quadratically-bounded losses on general data von Matus Telgarsky

Volltext

Spectrum Extraction and Clipping for Implicitly Linear Layers von Boroojeny, Ali Ebrahimpour, Telgarsky, Matus, Sundaram, Hari

Volltext bestellen

Transformers, parallel computation, and logarithmic depth von Sanford, Clayton, Hsu, Daniel, Telgarsky, Matus

Volltext bestellen

Neural networks and rational functions von Telgarsky, Matus

Volltext bestellen

Directional convergence and alignment in deep learning von Ji, Ziwei, Telgarsky, Matus

Volltext bestellen

Representational Strengths and Limitations of Transformers von Sanford, Clayton, Hsu, Daniel, Telgarsky, Matus

Volltext bestellen

Polylogarithmic width suffices for gradient descent to achieve arbitrarily small test error with shallow ReLU networks von Ji, Ziwei, Telgarsky, Matus

Volltext bestellen

Characterizing the implicit bias via a primal-dual analysis von Ji, Ziwei, Telgarsky, Matus

Volltext bestellen

Large Stepsize Gradient Descent for Logistic Loss: Non-Monotonicity of the Loss Improves Optimization Efficiency von Wu, Jingfeng, Bartlett, Peter L, Telgarsky, Matus, Yu, Bin

Volltext bestellen

Convex Analysis at Infinity: An Introduction to Astral Space von Dudík, Miroslav, Schapire, Robert E, Telgarsky, Matus

Volltext bestellen

Benefits of depth in neural networks von Telgarsky, Matus

Volltext bestellen

Representation Benefits of Deep Feedforward Networks von Telgarsky, Matus

Volltext bestellen

Size-Noise Tradeoffs in Generative Networks von Bailey, Bolton, Telgarsky, Matus

Volltext bestellen

On Achieving Optimal Adversarial Test Error von Li, Justin D, Matus Telgarsky

Volltext