#13 Project Mytheme

Notes toward a practical, interdisciplinary investigation. In simple terms: to build a ‘mytheme machine’, or a second-order signification system.

Read

#12 Chomsky: Basic Property

Notes on Chapter 1, ‘What is Language?’, from Noam Chomsky’s ‘What Kind of Creatures are We?’ (2016). Key focus: Basic Property of language as ‘language of thought’ (not communication).

Read

#11 Beyond Stochastic Parrots 🦜?

This entry introduces the debate emerging from two papers: ‘Emily Bender et al.’s ‘On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜’ and Steven T. Piantadosi, Felix Hill’s ‘Meaning without reference in large language models’.

Read

#10 Narrative and Number

Beginning with Shakespeare’s enigmatic stage direction, ‘Exit, pursued by a bear’, this entry refers to the ‘mathematics’ of narrative; to the seemingly infinite (yet finite?) combinatorial rules of language. As preface to the set reading on Roland Barthes, the entry considers again the fate of the statistical turn for AI and natural language processing.

Read

#9 Ut pictura poesis: Phoneme/Pixel

This entry offers two ‘exhibits’: the first is Elon Musk’s reference to ‘Wile E. Coyote stuff’ (in response to a question about adversarial attacks on a self-driving car); the other is Open AI’s DALL-E 2 system for creating images from natural language. Revealed is the enduring relationship between word and image.

Read

#8 Innate Structure

Is structuralism method or theory? In this entry Victor Burgin offers a translation of a conversation between Roman Jakobson, Claude Lévi-Strauss and geneticist Philippe L’Héritier, from a French broadcast in 1968.

Read

#7 Experimental Verification

In ‘The Structural Study of Myth’, Claude Lévi-Strauss offers an astonishingly compressed ‘formula’ for the complex study of myth: Fx(a): Fy(b) ≃ Fx(b): Fa-1(y). This is explored in relation to contemporary Non-Negative Matrix Factorization; the dominant algorithmic technique for astronomy, text mining, spectral data analysis, facial recognition, and more besides.

Read

#6 Anthropology / AGI

Commentary on Claude Lévi-Strauss’s collection of essays, The Other Side of the Moon, which this entry connects with considerations of artificial general intelligence and an analogy with astronomy; the imperfect way in which we schematise the stars, yet achieve significant regularities.

Read

#5 Shannon: Paper into Pixels

The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point. — Claude Shannon (1948)

Read

#4 Cybernetics / Responsible AI

As part of RLUK’s Digital Shift Forum, Ellen Broad, Senior Fellow, School of Cybernetics, Australian National University, and author of Made by Humans: The AI Condition, spoke on a cybernetic approach to responsible AI design.

Read

#3: Myth and Meaning

Preparatory notes for the first session of the structuralism reading group, which focuses on texts by Claude Lévi-Strauss. Entries on ‘Constituent Units’, ‘Invariance’, ‘Structuralism and the Subject’.

Read

#2: Atlas of AI

In a talk hosted by the Women in Data Science and AI project, part of The Alan Turing Institute’s Public Policy Programme, Kate Crawford discusses her excellent book, Atlas of AI (Yale University Press. 2021) with Judy Wajcman.

Read

#1: Initial Prompts

In preparing for the Turing Fellowship (2021-22), I drew upon three main prompts, which are set out here to help frame the discussions for the Reading Group. The first two prompts are specific and recent, while the third relates to a longer term interest.

Read