Douglas Meneghetti
AI specialist with a history in software engineering.
I am currently an AI Research Specialist at Samsung. Previously, I worked at Adroit Robotics, automating the monitoring of citrus orchards using computer vision techniques.
During my PhD, which I finished with a full scholarship by CAPES, I worked in the field of cooperative multi-agent reinforcement learnings. My research focused on the problem of fully cooperative multi-agent deep reinforcement learning with heterogeneous agents.
I was active for many years in robotics competitions (LARC and RoboCup) in the @Home and Humanoid Soccer categories, in which I focused in developing object detection systems for the robots.
I’ve also done professional work as a consultant and have publications in the area of educational assessment and computerized adaptive tests.
Would you like to know more?
I like to make stuff open for everybody. The things I build are kept in my GitHub repositories and the things I learn are kept in my blog. My projects page lists the things I’m proud of. ArtStation has my 3D art. StackExchange has my stupid questions and answers. Also check my publications.
Also check my links at the bottom of the page.
latest posts
selected publications
- CAPTION: Caption Analysis with Proposed Terms, Image of Objects, and Natural Language ProcessingSN Computer Science, Jul 2022
- Application and Simulation of Computerized Adaptive Tests Through the Package CatsimarXiv:1707.03012 [stat], Jul 2018
-
- Detecting Soccer Balls with Reduced Neural Networks: A Comparison of Multiple Architectures under Constrained Hardware ScenariosJournal of Intelligent & Robotic Systems, Mar 2021
- Specializing Inter-Agent Communication in Heterogeneous Multi-Agent Reinforcement Learning Using Agent Class InformationIn AAAI-21 Workshop on Reinforcement Learning in Games , Feb 2021