We invite submissions on novel research results (theoretical and empirical), software frameworks and abstractions, benchmarks, demos, visualizations, and work-in-progress research. The format of submissions is 4-page papers (excluding references) submitted to OpenReview. The reviews will not be shared publicly.
The papers at the intersection of physics, neuroscience, and biology will be considered for a Special Collection on Associative Memory and Hopfield Networks in PRX Life, a highly selective and fully open access journal focusing on topics at the interface of physics and biology. The publication fees will be waived for all accepted manuscripts in the Collection.
Associative memory is defined as a network that can link a set of features into high-dimensional vectors, called memories. Prompted by a large enough subset of features taken from one memory, an animal or an AI network with an associative memory can retrieve the rest of the features belonging to that memory. The diverse human cognitive abilities which involve making appropriate responses to stimulus patterns can often be understood as the operation of an associative memory, with the memories often being distillations and consolidations of multiple experiences rather than merely corresponding to a single event.
In the world of artificial neural networks, a canonical mathematical model of this phenomenon is the Hopfield network (Hopfield, 1982). Although often narrowly viewed as a model that can store and retrieve predefined verbatim memories of past events, its contemporary variants make it possible to store consolidated memories turning individual experiences into useful representations of the training data. Such modern variants are often trained using the backpropagation algorithm and often benefit from superior memory storage properties (Krotov & Hopfield, 2016). Contemporary Hopfield networks can be used as submodules in larger AI networks solving a diverse set of tasks. The goal of this workshop is to discuss the existing and emerging developments of these ideas. The research topics of interest at this workshop include (but are not limited to):
The OpenReview submission is now open: link
The submission deadline is: October 6 2023 23:59 AOE
September 29, 2023 23:59 AOE
Author notification by: October 26 2023 23:59 AOE
The format of submissions is 4-page papers (+ references) using the workshop's LaTeX template LaTeX template. Supplementary materials / appendices after the references are allowed and do not count towards the page limit. Submissions should be anonymized and should not include any identifying information about author identities or affiliations. There are no formal proceedings generated from this workshop. Accepted papers will be made public on OpenReview. The reviewing process will be double-blind. In assessing submitted contributions we will use the same rules for conflicts of interest as are used in the main track NeurIPS conference, e.g. reviewers cannot be from the same organization as authors, recent coauthors cannot review each other's submissions. While we welcome short versions of published papers, preference will be given to new and not yet published work.