Download This PaperPattern Prediction using LSTM and ReLU Ensemble
4 Pages Posted: 7 Apr 2020 Last revised: 7 Jul 2020
Date Written: April 6, 2020
Abstract
Human beings, like many other cognizant biological organisms, can roughly predict certain patterns in nature, be it with respect to time, or otherwise. In the strive to achieve the global object of achieving a general artificial intelligence, the authors deem it to be of high importance that the intelligence is aware of its surroundings, not just in terms of understanding raw sensory data, but also by learning and being able to predict recurring patterns in the environment. To achieve that result, this paper comes into play by discussing certain neural network models which help in learning from any sort of pattern recognition in a data sequence which can be achieved by either manual input or via data collection from the outer environment. This model is able to predict accurately about 80% of the time, beating state-of-the-art approaches.
Keywords: artificial intelligence, pattern detection, pattern recognition, machine learning, deep learning, rock paper scissors
Suggested Citation: Suggested Citation