Learning to make machine learning agents work on visual inputs has been a hurdle that researchers have faced since the early days of machine learning. There are currently techniques like deep learning and reinforcement learning which can be used in multi-feature environments. These techniques have stood their ground for a long time and have proven to be efficient. Therefore combining these fundamental concepts in order to realize a bigger goal is the best way to get best out of both. The Deep Learning Agent is to be designed using traditional machine learning methods like deep learning, reinforcement learning and deep Q-learning. Hence the agent is able to make the highest rewarding decision the will maximize the agent skill and make the learning process worth-while and efficient

Reinforcement Learning, Deep Learning, Game Bot, Agent, Action, Environment, Reward

[1] M.A.J.Bourassa and L.Massey, “Artificial Intelligence In Games A Survey Of The State Of The Art”, Defence Research &Development Organization Canada.(DRDC). Technical Memorandum, pp.3-40. DRDC Ottawa TM 2012-084 August, 2012.
[2] Robert Loftin, Michel.L.Littman, Jeff Huang, “A Strategy Aware Technique for Learning Behaviours from Discrete human feedback”, Berckely Brown Association for the Advancement of Artificial Intelligence, pp 1-5. June, 2014
[3] Jacob Schrum, Risto Miikkulakainen. “Constructing Game Agents through Simulated Evolution.”, In Encyclopedia of Computer Graphics and Games, pp.1-10 March, 2016 Springer.
[4] Jason Rennie, Andrew McCallum. “Using Reinforcement Learning To Spider The Web Efficiently”, ICML proceedings of sixteenth International Conference on Machine Learning, pp.334-345. June 27-30, 2009.
[5] Philip Hingston Senior Member IEEE. “A Turing Test For Computer Game Bots”, IEEE Transactions On Computational Intelligence in AI in Games, pp.1-18. September, 2009.
[6] Matthew, Peter Stone. “Deep Reinforcement Learning In Parameterized Action Space”, ICLR International Conference on Learning Representations, pp.143-155. Feb 2016.
[7] Igor. V. Karpov, Jacob Schkrum, Risto Miikkulakainen. In Philip.F.Hingston,“Belivable Bot Navigation via Playback of Human Traces", Believable Bots, pp.151-170. 2012. Springer.
[8] N.S.Lele, “Image Classification Using Convolutional Neural Network”. International Journal of Scientific Research in Computer Science and Engineering Vol.6, Issue.3, pp.22-26, June, 2018.
[9] A. Deepa, E. Chandra Blessie. “Input Analysis for Accreditation Prediction in Higher Education Sector by Using Gradient Boosting Algorithm”. International Journal of Scientific Research in Network Security and Communication. Vol-6, Issue-3, June, 2018.