In this paper, a multi-core RBM processor design with on-chip learning and reconfigurable sparsity is proposed to reduce energy consumption and improve processing throughput. The FPGA implementation results show that the proposed RBM design achieves 44.0% energy saving and 24.3% speed improvement in RBM training operation against the baseline CD-based RBM design. In the future, we will focus on ASIC implementation of our proposed RBM processor to further improve the energy efficiency and minimize the hardware cost.