The present invention relates to an Optimal Transport Curriculum Adaptive Learning System for Efficient Deep Visual Learning and method thereof, implemented as a structured computing system configured to dynamically regulate the progression of training data during neural network optimization. The system comprises a data intake unit, a feature representation processor, a distribution alignment processor, a curriculum sequencing unit, a neural training processor, a monitoring processor, a memory unit, and an interconnected communication arrangement that enables continuous adaptive interaction among the components. The system is configured to generate hierarchical feature representations from visual training samples and determine distributional relationships between an evolving representation state and a target training distribution. Based on computed alignment measures, the curriculum sequencing unit progressively regulates the order of presentation of training samples to maintain stable learning progression and balanced exposure to diverse visual feature distributions.