QuickNet: Maximizing Efficiency and Efficacy in Deep Architectures