A Collection of Conflicting Ideas

Welcome to my blog 馃憢馃徏.

  • This is Saeed Hedayatian. I’m a first year PhD student at University of Southern California. Currently, my main research focus is on multi-agent decision making + quality diversity optimization. More broadly, I’m interested in the connection between ideas from QD optimization and decision-making, whether in RL, MARL, recommender systems, etc. I also like to explore how techniques from RL can help design more powerful QD algorithms.
  • This blog is where I share intereting opinions, ideas, notes, etc. But mainly et cetera!

Variational Inference

Preliminaries It is usually the case that we have a dataset $\mathcal{D} = {x_1, \cdots, x_N}$ and a parametrized family of distributions $p_\theta (x)$. We would like to find the parameters that best describe the data. This is typically done using [[MLE and MAP|maximum likelihood estimation (MLE)]]. In this method, the optimal parameters are those that maximize the log likelihood of the data. Mathematically speaking, $$ \hat{\theta}_\mathrm{MLE} = \arg\max_\theta \frac{1}{N}\sum_{i=1}^{N}\log p_{\theta}(x_i)....

March 7, 2023 路 14 min 路 Saeed Hedayatian

In Praise of Einsum

This is a short note about the einsum functionality that is present in numpy, jax, etc. Understanding what it does is a bit tricky -naturally, because it can do the job of many other functions- but it is also very useful and can help a lot with linear algebraic computations. I will use numpy鈥檚 np.einsum() notation, but the underlaying concepts are the same regardless of syntactic differences in other libraries....

February 19, 2023 路 8 min 路 Saeed Hedayatian

MAP-Elites

MAP-Elites is an elegant algorithm for solving general optimization problems. To be more accurate, it is an illumination algorithm that tries to find high-performing and diverse solutions in a search space. At its core, it is a simple algorithm, both conceptually and to implement. Here, I briefly introduce the main idea behind the algorithm and its components. I will also discuss its merits and demerits compared to other approaches. This note is based on Illuminating Search Spaces by Mapping Elites....

September 29, 2022 路 6 min 路 Saeed Hedayatian

Learning to Score Behaviors

(This is a note based on Learning to Score Behaviors for Guided Policy Optimization. I am trying to expand and clarify some of the algorithms that were presented there. More content will be added to this note in the future!) The core question: What is the right measure of similarity between two policies acting on the same underlaying MDP and how can we devise algorithms to leverage this information for RL?...

September 21, 2022 路 9 min 路 Saeed Hedayatian

Optimization Primer

(Based on a lecture by professor Coralia Cartis, University of Oxford) (I don鈥檛 currently plan to extend it, but may expand and add more details to some of the later chapters in the future. I also like to eventually add some useful resources (books, talks, notes, etc.) about optimization) This brief note is about optimization problems. Though the main focus is on the general non-convex optimization problem, a lot of the methods borrow some ideas from convex optimization, so there are a lot of similarities....

August 27, 2022 路 31 min 路 Saeed Hedayatian