On the Inductive Bias of Stacking Towards Improving Reasoning von Saunshi, Nikunj, Karp, Stefani, Krishnan, Shankar, Miryoosefi, Sobhan, Reddi, Sashank J, Kumar, Sanjiv

Volltext bestellen

Landscape-Aware Growing: The Power of a Little LAG von Karp, Stefani, Saunshi, Nikunj, Miryoosefi, Sobhan, Reddi, Sashank J, Kumar, Sanjiv

Volltext bestellen

Private Adaptive Optimization with Side Information von Li, Tian, Zaheer, Manzil, Reddi, Sashank J, Smith, Virginia

Volltext bestellen

Depth Dependence of $\mu$P Learning Rates in ReLU MLPs von Jelassi, Samy, Hanin, Boris, Ji, Ziwei, Reddi, Sashank J, Bhojanapalli, Srinadh, Kumar, Sanjiv

Volltext bestellen

Differentially Private Adaptive Optimization with Delayed Preconditioners von Li, Tian, Zaheer, Manzil, Liu, Ken Ziyu, Reddi, Sashank J, McMahan, H. Brendan, Smith, Virginia

Volltext bestellen

Efficient Document Ranking with Learnable Late Interactions von Ji, Ziwei, Jain, Himanshu, Veit, Andreas, Reddi, Sashank J, Jayasumana, Sadeep, Rawat, Ankit Singh, Menon, Aditya Krishna, Yu, Felix, Kumar, Sanjiv

Volltext bestellen

On the Convergence of Adam and Beyond von Reddi, Sashank J, Kale, Satyen, Kumar, Sanjiv

Volltext bestellen

Robust Training of Neural Networks Using Scale Invariant Architectures von Li, Zhiyuan, Bhojanapalli, Srinadh, Zaheer, Manzil, Reddi, Sashank J, Kumar, Sanjiv

Volltext bestellen

Distilling Double Descent von Cotter, Andrew, Menon, Aditya Krishna, Narasimhan, Harikrishna, Rawat, Ankit Singh, Reddi, Sashank J, Zhou, Yichen

Volltext bestellen

On the Role of Depth and Looping for In-Context Learning with Task Diversity von Gatmiry, Khashayar, Saunshi, Nikunj, Reddi, Sashank J, Jegelka, Stefanie, Kumar, Sanjiv

Volltext

Can Looped Transformers Learn to Implement Multi-step Gradient Descent for In-context Learning? von Gatmiry, Khashayar, Saunshi, Nikunj, Reddi, Sashank J, Jegelka, Stefanie, Kumar, Sanjiv

Volltext

On the Inductive Bias of Stacking Towards Improving Reasoning von Saunshi, Nikunj, Karp, Stefani, Krishnan, Shankar, Sobhan Miryoosefi, Reddi, Sashank J, Kumar, Sanjiv

Volltext

On the Algorithmic Stability and Generalization of Adaptive Optimization Methods von Nguyen, Han, Pham, Hai, Reddi, Sashank J, Póczos, Barnabás

Volltext

O(n)$ Connections are Expressive Enough: Universal Approximability of Sparse Transformers von Yun, Chulhee, Chang, Yin-Wen, Bhojanapalli, Srinadh, Rawat, Ankit Singh, Reddi, Sashank J, Kumar, Sanjiv

Volltext bestellen

Landscape-Aware Growing: The Power of a Little LAG von Karp, Stefani, Saunshi, Nikunj, Sobhan Miryoosefi, Reddi, Sashank J, Kumar, Sanjiv

Volltext

Doubly-stochastic mining for heterogeneous retrieval von Rawat, Ankit Singh, Menon, Aditya Krishna, Veit, Andreas, Yu, Felix, Reddi, Sashank J, Kumar, Sanjiv

Volltext bestellen

Adaptive Sampling Distributed Stochastic Variance Reduced Gradient for Heterogeneous Distributed Datasets von Ramazanli, Ilqar, Nguyen, Han, Pham, Hai, Reddi, Sashank J, Poczos, Barnabas

Volltext bestellen

Are Transformers universal approximators of sequence-to-sequence functions? von Yun, Chulhee, Bhojanapalli, Srinadh, Rawat, Ankit Singh, Reddi, Sashank J, Kumar, Sanjiv

Volltext bestellen

Why are Adaptive Methods Good for Attention Models? von Zhang, Jingzhao, Karimireddy, Sai Praneeth, Veit, Andreas, Kim, Seungyeon, Reddi, Sashank J, Kumar, Sanjiv, Sra, Suvrit

Volltext bestellen

Riemannian SVRG: Fast Stochastic Optimization on Riemannian Manifolds von Zhang, Hongyi, Reddi, Sashank J, Sra, Suvrit

Volltext bestellen