Hadi Daneshmand

Publication Activity (10 Years)

Years Active: 2014-2024
Publications (10 Years): 32

Top Topics

Temporal Difference

Sample Complexity

Pac Learnability

Top Venues

Publications

Alexandru Meterez, Amir Joudaki, Francesco Orabona, Alexander Immer, Gunnar Rätsch, Hadi Daneshmand
Towards Training Without Depth Limits: Batch Normalization Without Gradient Explosion. ICLR (2024)
Jiuqi Wang, Ethan Blaser, Hadi Daneshmand, Shangtong Zhang
Transformers Learn Temporal Difference Methods for In-Context Reinforcement Learning. CoRR (2024)
Kwangjun Ahn, Xiang Cheng, Hadi Daneshmand, Suvrit Sra
Transformers learn to implement preconditioned gradient descent for in-context learning. CoRR (2023)
Alexandru Meterez, Amir Joudaki, Francesco Orabona, Alexander Immer, Gunnar Rätsch, Hadi Daneshmand
Towards Training Without Depth Limits: Batch Normalization Without Gradient Explosion. CoRR (2023)
Amir Joudaki, Hadi Daneshmand, Francis R. Bach
On the impact of activation and normalization in obtaining isometric embeddings at initialization. CoRR (2023)
Kwangjun Ahn, Xiang Cheng, Hadi Daneshmand, Suvrit Sra
Transformers learn to implement preconditioned gradient descent for in-context learning. NeurIPS (2023)
Hadi Daneshmand, Jason D. Lee, Chi Jin
Efficient displacement convex optimization with particle gradient descent. CoRR (2023)
Amir Joudaki, Hadi Daneshmand, Francis R. Bach
On the impact of activation and normalization in obtaining isometric embeddings at initialization. NeurIPS (2023)
Hadi Daneshmand, Jason D. Lee, Chi Jin
Efficient displacement convex optimization with particle gradient descent. ICML (2023)
Amir Joudaki, Hadi Daneshmand, Francis R. Bach
On Bridging the Gap between Mean Field and Finite Width Deep Random Multilayer Perceptron with Batch Normalization. ICML (2023)
Hadi Daneshmand, Francis R. Bach
Polynomial-time sparse measure recovery. CoRR (2022)
Amir Joudaki, Hadi Daneshmand, Francis R. Bach
Entropy Maximization with Depth: A Variational Principle for Random Neural Networks. CoRR (2022)
Hadi Daneshmand, Amir Joudaki, Francis R. Bach
Batch Normalization Orthogonalizes Representations in Deep Random Networks. NeurIPS (2021)
Peiyuan Zhang, Antonio Orvieto, Hadi Daneshmand
Rethinking the Variational Interpretation of Accelerated Optimization Methods. NeurIPS (2021)
Hadi Daneshmand, Amir Joudaki, Francis R. Bach
Batch Normalization Orthogonalizes Representations in Deep Random Networks. CoRR (2021)
Peiyuan Zhang, Antonio Orvieto, Hadi Daneshmand, Thomas Hofmann, Roy S. Smith
Revisiting the Role of Euler Numerical Integration on Acceleration and Stability in Convex Optimization. AISTATS (2021)
Peiyuan Zhang, Antonio Orvieto, Hadi Daneshmand, Thomas Hofmann, Roy S. Smith
Revisiting the Role of Euler Numerical Integration on Acceleration and Stability in Convex Optimization. CoRR (2021)
Hadi Daneshmand, Jonas Moritz Kohler, Francis R. Bach, Thomas Hofmann, Aurélien Lucchi
Batch normalization provably avoids ranks collapse for randomly initialised deep networks. NeurIPS (2020)
Hadi Daneshmand, Jonas Moritz Kohler, Francis R. Bach, Thomas Hofmann, Aurélien Lucchi
Theoretical Understanding of Batch-normalization: A Markov Chain Perspective. CoRR (2020)
Leonard Adolphs, Hadi Daneshmand, Aurélien Lucchi, Thomas Hofmann
Local Saddle Point Optimization: A Curvature Exploitation Approach. AISTATS (2019)
Peiyuan Zhang, Hadi Daneshmand, Thomas Hofmann
Mixing of Stochastic Accelerated Gradient Descent. CoRR (2019)
Jonas Moritz Kohler, Hadi Daneshmand, Aurélien Lucchi, Thomas Hofmann, Ming Zhou, Klaus Neymeyr
Exponential convergence rates for Batch Normalization: The power of length-direction decoupling in non-convex optimization. AISTATS (2019)
Jonas Moritz Kohler, Hadi Daneshmand, Aurélien Lucchi, Ming Zhou, Klaus Neymeyr, Thomas Hofmann
Towards a Theoretical Understanding of Batch Normalization. CoRR (2018)
Hadi Daneshmand, Jonas Moritz Kohler, Aurélien Lucchi, Thomas Hofmann
Escaping Saddles with Stochastic Gradients. CoRR (2018)
Hadi Daneshmand, Jonas Moritz Kohler, Aurélien Lucchi, Thomas Hofmann
Escaping Saddles with Stochastic Gradients. ICML (2018)
Leonard Adolphs, Hadi Daneshmand, Aurélien Lucchi, Thomas Hofmann
Local Saddle Point Optimization: A Curvature Exploitation Approach. CoRR (2018)
Hadi Daneshmand, Hamed Hassani, Thomas Hofmann
Accelerated Dual Learning by Homotopic Initialization. CoRR (2017)
Hadi Daneshmand, Aurélien Lucchi, Thomas Hofmann
Starting Small - Learning with Adaptive Sample Sizes. ICML (2016)
Hadi Daneshmand, Aurélien Lucchi, Thomas Hofmann
Starting Small - Learning with Adaptive Sample Sizes. CoRR (2016)
Manuel Gomez-Rodriguez, Le Song, Hadi Daneshmand, Bernhard Schölkopf
Estimating Diffusion Networks: Recovery Conditions, Sample Complexity and Soft-thresholding Algorithm. J. Mach. Learn. Res. 17 (2016)
Aryan Mokhtari, Hadi Daneshmand, Aurélien Lucchi, Thomas Hofmann, Alejandro Ribeiro
Adaptive Newton Method for Empirical Risk Minimization to Statistical Accuracy. NIPS (2016)
Hadi Daneshmand, Aurélien Lucchi, Thomas Hofmann
DynaNewton - Accelerating Newton's Method for Machine Learning. CoRR (2016)
Hadi Daneshmand, Manuel Gomez-Rodriguez, Le Song, Bernhard Schölkopf
Estimating Diffusion Network Structures: Recovery Conditions, Sample Complexity & Soft-thresholding Algorithm. CoRR (2014)
Hadi Daneshmand, Manuel Gomez-Rodriguez, Le Song, Bernhard Schölkopf
Estimating Diffusion Network Structures: Recovery Conditions, Sample Complexity & Soft-thresholding Algorithm. ICML (2014)