Mendeley advisor of the month: Gabriel de Oliveira Ramos is a postdoctoral research fellow at the Artificial Intelligence Lab from the Vrije Universiteit Brussel (Belgium). He obtained his PhD (with highest honours) and MSc degrees in Computer Science from the Federal University of Rio Grande do Sul (Brazil) in 2018 and 2013, respectively. Ramos’ research focuses on multiagent reinforcement learning and game theory, especially in the context of complex scenarios, such as traffic and smart grids.

How did you get into your field and what is your research story?

I started to write my first computer programs at 14 and developed, since then, my passion for Computer Science. Not much later, during my bachelor’s first year, I got in contact with Artificial Intelligence (AI) for the first time and decided that AI would be my research field. In the following years, I developed my research on different AI topics, including machine learning, game theory, and planning. In all cases, my research has always been motivated by real-world problems, like traffic, electricity grids, and logistics. Moreover, the theoretical properties of my methods have always played a role in my research.

Where do you do your research/work the best? What kind of environment suits you?

Any environment where I can balance insightful discussion sessions (with my peers) with silent study sessions. Good computer resources are also extremely useful, together with the traditional paper-and-pen combination.

How long have you been on Mendeley? 

I started using Mendeley in June 2013, just after I finished my masters, to organize the mess of my references at that time.

What were you using prior to Mendeley and how does Mendeley influence your research?

Keeping track of the literature is fundamental in science. Before using Mendeley, I had all my references grouped by topics into folders of my computer. The main problem, however, was to efficiently store my annotations and conclusions about such references. With Mendeley, I could finally store all my notes in an efficient and reliable way. Together with the nice search mechanism, it became easier for me to focus on my research.

Why did you decide to become an Advisor and how are you involved with the program?

I have been a Mendeley enthusiast since I started using it (indeed, it has considerably increased my productivity on specific tasks). As such, I always spread the word about it. Moreover, I contributed to Mendeley by suggesting important improvement several times. In this sense, I always felt as an informal advisor, which became a formal status in May 2015.

What researcher would you like to work with or meet, dead or alive?

My work has been inspired by so many brilliant researchers that I could not mention all of them here. Among them, I should definitely highlight Prof. Avrim Blum (TTI-Chicago), Prof. Michael Bowling (UAlberta), and Prof. Tim Roughgarden (Stanford), whose works motivated (and shed light on) my PhD research.

What book are you reading at the moment and why?

The second edition of Reinforcement Learning: An Introduction, by Sutton and Barto. It is always important to refresh such fundamental topics.

What’s the most interesting thing you’ve learned this week?

I attended some of the world’s most important conferences on Artificial Intelligence (ICML and AAMAS), and I enforced to myself the belief that, as a researcher, you should always be open-minded and eager for learning new things.

What is the best part about working in research?

You are always learning new ways of solving problems that could potentially improve people’s lives.

And the worst/most challenging part about working in research?

Sometimes (almost always, in fact) the answer is not the one you would expect. Although challenging, that is what moves science forward (and actually, that is one of the most exciting parts of doing science).

What is the one thing you want people to know about Mendeley?

Mendeley really makes your reference management easier.