Melika Emami

Ph.D. Candidate, Department of Electrical and Computer Engineering, University of California, Los Angeles (UCLA). Contact: emami [at] ucla [dot] edu Google Scholar LinkedIn GitHub

About Me

I am a Ph.D. candidate in the ECE department at UCLA working under the supervision of Prof. Alyson Fletcher and Prof. Sundeep Rangan. I received my Master’s degree at UCLA in 2018, and my Bachelor's degree at University of Tehran in 2016, both in Electrical and Computer Engineering.

Research Interests

My research is focused on theoretical machine learning with the broad goal of understanding the asymptotics of learning in deep neural networks. I apply tools from high-dimensional statistics and large-scale optimization to provide rigorous theoretical guarantees for learning and generalization in machine learning problems. I am also interested in high-dimensional time series models with applications to computational neuroscience such as modeling complex neural sensory responses in the brain.

News

• May 2022: I will give a talk in the Math Machine Learning seminar at the Max Planck Institute for Mathematics in the Sciences and UCLA (MPI MiS + UCLA). [video and abstract]

• January 2022: New preprint available on arXiv: Kernel Methods and Multi-layer Perceptrons Learn Linear Models in High Dimensionss

• Decemmber 2021: New preprint available on arXiv: Augmented Contrastive Self-Supervised Learning for Audio Invariant Representations

• July 2021: Presenting our paper at ICML 2021. [video]

• May 2021: Paper accepted at ICML 2021: Implicit Bias of Linear RNNs.

• March 2021: I will join Microsoft as a Research Intern in Spring 2021.

• March 2021: I am a Ph.D Candidate now.

• January 2021: New preprint available on arXiv: Implicit Bias of Linear RNNs

• July 2020: Presenting our paper at ICML 2020. [video] [slides]

• May 2020: Paper accepted at ICML 2020: Generalization Error of Generalized Linear Models in High Dimensions

• February 2020: Presenting our recent work at ITA 2020, San Diego, CA.

• October 2019: Presenting our recent work at DeepMath 2019, New York, NY. [poster]

• September 2019: Paper accepted at NeurIPS 2019: Input-Output Equivalence of Unitary and Contractive RNNs

• November 2018: Presenting our recent work at Society for Neuroscience Meeting 2018, San Diego, CA. [poster]

• September 2018: Presenting our paper at CCN, philadelphia, PA: Low-Rank Nonlinear Decoding of Micro-ECoG from the Primary Auditory Cortex