Most physical realizations of a quantum computer are based on a linear nearest neighbor (LNN) arrangement of qubits where qubits are arranged along an array, and only the adjacent qubits interact with each other. To perform an operation between two non-NN (remote) qubits, additional gates are required to bring the qubits adjacent to each other, which can increase the computational overhead. Here, we propose a new scheme using a learning algorithm as a tool, to find the system parameters that perform Controlled-Not (CNOT) gate efficiently between remote qubits in LNN architectures which is an elementary operation in quantum systems.