Alert button
Picture for Mahdi Soltanolkotabi

Mahdi Soltanolkotabi

Alert button

On the Role of Attention in Prompt-tuning

Jun 06, 2023
Samet Oymak, Ankit Singh Rawat, Mahdi Soltanolkotabi, Christos Thrampoulidis

Figure 1 for On the Role of Attention in Prompt-tuning
Figure 2 for On the Role of Attention in Prompt-tuning
Figure 3 for On the Role of Attention in Prompt-tuning
Figure 4 for On the Role of Attention in Prompt-tuning
Viaarxiv icon

DiracDiffusion: Denoising and Incremental Reconstruction with Assured Data-Consistency

Mar 25, 2023
Zalan Fabian, Berk Tinaz, Mahdi Soltanolkotabi

Figure 1 for DiracDiffusion: Denoising and Incremental Reconstruction with Assured Data-Consistency
Figure 2 for DiracDiffusion: Denoising and Incremental Reconstruction with Assured Data-Consistency
Figure 3 for DiracDiffusion: Denoising and Incremental Reconstruction with Assured Data-Consistency
Figure 4 for DiracDiffusion: Denoising and Incremental Reconstruction with Assured Data-Consistency
Viaarxiv icon

Implicit Balancing and Regularization: Generalization and Convergence Guarantees for Overparameterized Asymmetric Matrix Sensing

Mar 24, 2023
Mahdi Soltanolkotabi, Dominik Stöger, Changzhi Xie

Viaarxiv icon

CUDA: Convolution-based Unlearnable Datasets

Mar 07, 2023
Vinu Sankar Sadasivan, Mahdi Soltanolkotabi, Soheil Feizi

Figure 1 for CUDA: Convolution-based Unlearnable Datasets
Figure 2 for CUDA: Convolution-based Unlearnable Datasets
Figure 3 for CUDA: Convolution-based Unlearnable Datasets
Figure 4 for CUDA: Convolution-based Unlearnable Datasets
Viaarxiv icon

SMILE: Scaling Mixture-of-Experts with Efficient Bi-level Routing

Dec 10, 2022
Chaoyang He, Shuai Zheng, Aston Zhang, George Karypis, Trishul Chilimbi, Mahdi Soltanolkotabi, Salman Avestimehr

Figure 1 for SMILE: Scaling Mixture-of-Experts with Efficient Bi-level Routing
Figure 2 for SMILE: Scaling Mixture-of-Experts with Efficient Bi-level Routing
Figure 3 for SMILE: Scaling Mixture-of-Experts with Efficient Bi-level Routing
Figure 4 for SMILE: Scaling Mixture-of-Experts with Efficient Bi-level Routing
Viaarxiv icon

The Geometry of Self-supervised Learning Models and its Impact on Transfer Learning

Sep 18, 2022
Romain Cosentino, Sarath Shekkizhar, Mahdi Soltanolkotabi, Salman Avestimehr, Antonio Ortega

Figure 1 for The Geometry of Self-supervised Learning Models and its Impact on Transfer Learning
Figure 2 for The Geometry of Self-supervised Learning Models and its Impact on Transfer Learning
Figure 3 for The Geometry of Self-supervised Learning Models and its Impact on Transfer Learning
Figure 4 for The Geometry of Self-supervised Learning Models and its Impact on Transfer Learning
Viaarxiv icon

Neural Networks can Learn Representations with Gradient Descent

Jun 30, 2022
Alex Damian, Jason D. Lee, Mahdi Soltanolkotabi

Figure 1 for Neural Networks can Learn Representations with Gradient Descent
Figure 2 for Neural Networks can Learn Representations with Gradient Descent
Viaarxiv icon

Toward a Geometrical Understanding of Self-supervised Contrastive Learning

May 13, 2022
Romain Cosentino, Anirvan Sengupta, Salman Avestimehr, Mahdi Soltanolkotabi, Antonio Ortega, Ted Willke, Mariano Tepper

Figure 1 for Toward a Geometrical Understanding of Self-supervised Contrastive Learning
Figure 2 for Toward a Geometrical Understanding of Self-supervised Contrastive Learning
Figure 3 for Toward a Geometrical Understanding of Self-supervised Contrastive Learning
Figure 4 for Toward a Geometrical Understanding of Self-supervised Contrastive Learning
Viaarxiv icon

Learning from many trajectories

Mar 31, 2022
Stephen Tu, Roy Frostig, Mahdi Soltanolkotabi

Figure 1 for Learning from many trajectories
Figure 2 for Learning from many trajectories
Figure 3 for Learning from many trajectories
Viaarxiv icon

HUMUS-Net: Hybrid unrolled multi-scale network architecture for accelerated MRI reconstruction

Mar 15, 2022
Zalan Fabian, Mahdi Soltanolkotabi

Figure 1 for HUMUS-Net: Hybrid unrolled multi-scale network architecture for accelerated MRI reconstruction
Figure 2 for HUMUS-Net: Hybrid unrolled multi-scale network architecture for accelerated MRI reconstruction
Figure 3 for HUMUS-Net: Hybrid unrolled multi-scale network architecture for accelerated MRI reconstruction
Figure 4 for HUMUS-Net: Hybrid unrolled multi-scale network architecture for accelerated MRI reconstruction
Viaarxiv icon