English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Neural-symbolic approaches to machine learning incorporate the advantagesfrom both connectionist and symbolic methods. Typically, these models employ afirst module based on a neural architecture to extract features from complexdata. Then, these features are processed as symbols by a symbolic engine thatprovides reasoning, concept structures, composability, better generalizationand out-of-distribution learning among other possibilities. However, neuralapproaches to the grounding of symbols in sensory data, albeit powerful, stillrequire heavy training and tedious labeling for the most part. This paperpresents a new symbolic-only method for the generation of hierarchical conceptstructures from complex spatial sensory data. The approach is based onBateson's notion of difference as the key to the genesis of an idea or aconcept. Following his suggestion, the model extracts atomic features from rawdata by computing elemental sequential comparisons in a stream of multivariatenumerical values. Higher-level constructs are built from these features bysubjecting them to further comparisons in a recursive process. At any stage inthe recursion, a concept structure may be obtained from these constructs andfeatures by means of Formal Concept Analysis. Results show that the model isable to produce fairly rich yet human-readable conceptual representationswithout training. Additionally, the concept structures obtained through themodel (i) present high composability, which potentially enables the generationof 'unseen' concepts, (ii) allow formal reasoning, and (iii) have inherentabilities for generalization and out-of-distribution learning. Consequently,this method may offer an interesting angle to current neural-symbolic research.Future work is required to develop a training methodology so that the model canbe tested against a larger dataset.

A Recursive Bateson-Inspired Model for the Generation of Semantic Formal Concepts from Spatial Sensory Data