Memory represents the past and the present experiences of a person. The storage and retrieval of these experiences are crucial for the survival of any living being. This storage and retrieval is supported by four key memory processes (encoding, retrieval, replay, and consolidation). The hippocampus plays a major role in all these memory processes. We have developed models of the hippocampus that explain these memory processes. The encoding process allows the formation of meaningful representations of the observed sensory information. The encoding and recall processes of the hippocampus are explained using associative memory using deep learning and reinforcement learning approaches. The proposed model also mimics the characteristics of Alzheimer’s patients in the picture-naming task. The memory consolidation process enables the accumulation of recent memory along with the remote memories in the long-term memory store. In general, the deep network models of memory suffer from forgetting old information while learning new information, called catastrophic forgetting/interference. The human brain overcomes this problem quite effectively, a problem that continues to challenge current deep neural network models. We propose a regularization-based model to solve the problem of catastrophic forgetting. Hippocampal place cells demonstrate neuronal replay of behavioral sequences at a faster timescale in forward and reverse directions during resting states (awake and sleep). We propose a model of the hippocampus to demonstrate replay characteristics using a hybrid model that uses an oscillator and deep value networks.