A cognitive beamforming algorithm for colocated MIMO radars, based on Reinforcement Learning (RL) framework, is proposed. We analyse an RL-based optimization protocol that allows the MIMO radar, i.e. the \textit{agent}, to iteratively sense the unknown environment, i.e. the radar scene involving an unknown number of targets at unknown angular positions, and consequently, to synthesize a set of transmitted waveforms whose related beam patter is tailored on the acquired knowledge. The performance of the proposed RL-based beamforming algorithm is assessed through numerical simulations in terms of Probability of Detection ($P_D$).