Reverse Engineering Creativity into Interpretable Neural Networks

Oita, Marilena

doi:10.1007/978-3-030-12385-7_19

Cited by 3 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This problem of interpretability and explainability of how or why the DDN has made some set of connections is not trivial, and makes it difficult for researchers to traditionally understand what is being encoded on each neuron in each layer of the DNN ( Towell and Shavlik, 1993 ; Liu X. et al, 2018 ; Kumar et al, 2020 ; Erasmus et al, 2021 ). This ability to understand some of the encodings can be important with applications such as medical decision making, law enforcement, financial analysis ( Horta et al, 2021 ) as well as when attempting to model and explain the cognitive system ( Cichy and Kaiser, 2019 ; Oita, 2019 ; Monte-Serrat and Cattani, 2021 ) such as in tasks relating to background knowledge which my further help researchers understand distributed knowledge encodings across layers and what the weight distributions actually mean in graphical form.…”

Section: Encoding Of Information To Network Layers and Graph Visual E...mentioning

confidence: 99%

A Functional Contextual Account of Background Knowledge in Categorization: Implications for Artificial General Intelligence and Cognitive Accounts of General Knowledge

2022

View full text Add to dashboard Cite

Psychology has benefited from an enormous wealth of knowledge about processes of cognition in relation to how the brain organizes information. Within the categorization literature, this behavior is often explained through theories of memory construction called exemplar theory and prototype theory which are typically based on similarity or rule functions as explanations of how categories emerge. Although these theories work well at modeling highly controlled stimuli in laboratory settings, they often perform less well outside of these settings, such as explaining the emergence of background knowledge processes. In order to explain background knowledge, we present a non-similarity-based post-Skinnerian theory of human language called Relational Frame Theory (RFT) which is rooted in a philosophical world view called functional contextualism (FC). This theory offers a very different interpretation of how categories emerge through the functions of behavior and through contextual cues, which may be of some benefit to existing categorization theories. Specifically, RFT may be able to offer a novel explanation of how background knowledge arises, and we provide some mathematical considerations in order to identify a formal model. Finally, we discuss much of this work within the broader context of general semantic knowledge and artificial intelligence research.

show abstract