As billions of devices are connecting to the Internet, spectrum crunch has become a major concern for all mobile network operators. One of the ways to address the issue of lack of bandwidth is by sharing licensed spectrum among the operators. In this paper, we analyze handover probability, coverage probability, and data rate achieved in spectrum sharing. Furthermore, the increase in density of Base Stations (BSs) has resulted in increased energy consumption and increased interference to the cell-edge users. The increased energy consumption results in an increased operational costs while inefficient handling of cell-edge users results in decreased Quality of Experience (QoE) of users. Therefore, in this paper, we propose an energy efficient Q-Learning based framework called as Learning based Discrete Breathing which minimizes the energy consumption of the system without trading off throughput and also addresses interference to the cell-edge users. We analyze the proposed framework and by extensive simulation show the effectiveness of our proposed framework in terms of energy consumption, throughput, and energy consumption rating. © 2017 IEEE.