Throughout my Ph.D., my research was focused on advancing Generalized Zero-Shot Learning (GZSL) for Visual Object Recognition. I devised two cutting-edge generative techniques: one based on novel generative algorithms and another utilizing a VAE-based network for disentangled representation learning. These methodologies were applied to both single-label and multi-label ZSL problems, pioneering the study of disentangled representations in these intricate research areas.