Risultati della ricerca - Soudry, Daniel

PLUMAGE: Probabilistic Low rank Unbiased Min Variance Gradient Estimator for Efficient Large Model Training di Haroush, Matan, Soudry, Daniel

Testo

Alias-Free ViT: Fractional Shift Invariance via Linear Attention di Michaeli, Hagay, Soudry, Daniel

Testo

Alias-Free Convnets: Fractional Shift Invariance via Polynomial Activations di Michaeli, Hagay, Michaeli, Tomer, Soudry, Daniel

Testo

Towards Cheaper Inference in Deep Networks with Lower Bit-Width Accumulators di Blumenfeld, Yaniv, Hubara, Itay, Soudry, Daniel

Testo

Minimum Variance Unbiased N:M Sparsity for the Neural Gradients di Chmiel, Brian, Hubara, Itay, Banner, Ron, Soudry, Daniel

Testo

Explore to Generalize in Zero-Shot RL di Zisselman, Ev, Lavie, Itai, Soudry, Daniel, Tamar, Aviv

Testo

Scaling FP8 training to trillion-token LLMs di Fishman, Maxim, Chmiel, Brian, Banner, Ron, Soudry, Daniel

Testo

Foldable SuperNets: Scalable Merging of Transformers with Different Initializations and Tasks di Kinderman, Edan, Hubara, Itay, Maron, Haggai, Soudry, Daniel

Testo

The Implicit Bias of Gradient Descent on Separable Multiclass Data di Ravi, Hrithik, Scott, Clayton, Soudry, Daniel, Wang, Yutong

Testo

FP4 All the Way: Fully Quantized Training of LLMs di Chmiel, Brian, Fishman, Maxim, Banner, Ron, Soudry, Daniel

Testo

Exponential Quantum Communication Advantage in Distributed Inference and Learning di Gilboa, Dar, Michaeli, Hagay, Soudry, Daniel, McClean, Jarrod R.

Testo

How do Minimum-Norm Shallow Denoisers Look in Function Space? di Zeno, Chen, Ongie, Greg, Blumenfeld, Yaniv, Weinberger, Nir, Soudry, Daniel

Testo

The Joint Effect of Task Similarity and Overparameterization on Catastrophic Forgetting -- An Analytical Model di Goldfarb, Daniel, Evron, Itay, Weinberger, Nir, Soudry, Daniel, Hand, Paul

Testo

Temperature is All You Need for Generalization in Langevin Dynamics and other Markov Processes di Harel, Itamar, Wolanowsky, Yonathan, Vardi, Gal, Srebro, Nathan, Soudry, Daniel

Testo

Tensor-Parallelism with Partially Synchronized Activations di Lamprecht, Itay, Karnieli, Asaf, Hanani, Yair, Giladi, Niv, Soudry, Daniel

Testo

Normalized Architectures are Natively 4-Bit di Fishman, Maxim, Chmiel, Brian, Banner, Ron, Soudry, Daniel, Ginsburg, Boris

Testo

Workspace Optimization: How to Train Your Agent di Sarafian, Elad, Kaplun, Gal, Banner, Ron, Soudry, Daniel, Ginsburg, Boris

Testo

The Implicit Bias of Gradient Descent on Separable Data di Soudry, Daniel, Hoffer, Elad, Nacson, Mor Shpigel, Gunasekar, Suriya, Srebro, Nathan

Testo

Accurate Neural Training with 4-bit Matrix Multiplications at Standard Formats di Chmiel, Brian, Banner, Ron, Hoffer, Elad, Yaacov, Hilla Ben, Soudry, Daniel

Testo

Stable Minima Cannot Overfit in Univariate ReLU Networks: Generalization by Large Step Sizes di Qiao, Dan, Zhang, Kaiqi, Singh, Esha, Soudry, Daniel, Wang, Yu-Xiang

Testo