Latent Structured Hopfield Network for Semantic Association and Retrieval
This work addresses the challenge of episodic memory formation in AI, offering a biologically grounded computational model that could enhance memory retrieval in systems, though it appears incremental in building on existing Hopfield and autoencoder methods.
The paper tackled the problem of forming associative structures for episodic memory by proposing the Latent Structured Hopfield Network (LSHN), a biologically inspired framework that integrates Hopfield attractor dynamics into an autoencoder, and demonstrated superior performance in recalling corrupted inputs on datasets like MNIST and CIFAR-10, outperforming existing models.
Episodic memory enables humans to recall past experiences by associating semantic elements such as objects, locations, and time into coherent event representations. While large pretrained models have shown remarkable progress in modeling semantic memory, the mechanisms for forming associative structures that support episodic memory remain underexplored. Inspired by hippocampal CA3 dynamics and its role in associative memory, we propose the Latent Structured Hopfield Network (LSHN), a biologically inspired framework that integrates continuous Hopfield attractor dynamics into an autoencoder architecture. LSHN mimics the cortical-hippocampal pathway: a semantic encoder extracts compact latent representations, a latent Hopfield network performs associative refinement through attractor convergence, and a decoder reconstructs perceptual input. Unlike traditional Hopfield networks, our model is trained end-to-end with gradient descent, achieving scalable and robust memory retrieval. Experiments on MNIST, CIFAR-10, and a simulated episodic memory task demonstrate superior performance in recalling corrupted inputs under occlusion and noise, outperforming existing associative memory models. Our work provides a computational perspective on how semantic elements can be dynamically bound into episodic memory traces through biologically grounded attractor mechanisms. Code: https://github.com/fudan-birlab/LSHN.