Colin Acker Email & Phone Number

Atlanta, Ga, Us

Eglin Air Force Base, Florida, United States

Expanded on my work I did last summer. I am currently working on using multi-agent reinforcement learning to solve a variety of tasks. Such tasks include obstacle avoidance, pursuit, and evasion. Also, applying computer vision techniques such as convolutional neural networks and vision transformers to reinforcement learning environments. Developed a custom environment, using Pygame for rendering and Pymunk for physics. Used this environment for model-free, multi-agent reinforcement learning.

Florida, United States

I worked with graduate students and full time researchers in studying model-free reinforcement learning (RL) algorithms. I created custom multi-agent environments that are used for testing different RL algorithms. I proved that multiple agents within a continuous and complex state space can cooperate to avoid obstacles and reach a fleeing target using model-free algorithms, which no previous literature has been successful in doing. I worked with policy gradient methods, deep neural networks (DNNs), image processing neural networks (CNNs), PyTorch, OpenAI gym, and PyGame.

Florida, United States

I worked with graduate students and researchers in researching task assignment in a stochastic environment. I took previous research from the early 2000's, implemented their algorithms and environment, and expanded upon it. I created a more complex environment that was probabilistic and used path planning. I also used model-based and model-free RL algorithms to benchmark how they would preform on my model.

Douglasville, Georgia

Bachelor Of Science - Bs, Mathematics, Junior

High School Diploma, High School/Secondary Diplomas And Certificates