In the search for artificial general intelligence, model development and training has focused primarily on vast datasets of known problems and their accepted solutions. This process necessarily produces convergent systems which are fundamentally incapable of the conceptual reframing that is required for genuine creative breakthroughs. Inspired by the divergent cognitive processes that allow humans to make such creative leaps, our work introduces a family of language models, TinyTim, to serve as sources of divergent generation within broader systems. These models have been created by fine-tuning on the anti-parsimonious text of James Joyce's `Finnegans Wake'. Quantitative analysis of both an unsupervised fine-tuned model (TinyTim-V1) and a new instruction-tuned variant (TinyTim-V2) demonstrates a profound capacity for lexical invention; the foundational V1 model exhibits a Yule's K score for lexical richness over twenty times greater than that of convergent baselines. This trait is a stable property of the family, as the instruction-tuned V2 maintains a statistically distinct profile and resists factual convergence, sacrificing benchmark performance to preserve its core generative style. This work establishes a methodology for engineering specialized divergent models that, when paired with convergent systems, can reframe problems and force breakthroughs beyond the reach of statistical optimization alone.

The complexity characteristics of texts written in natural languages are significantly related to the rules of punctuation. In particular, the distances between punctuation marks measured by the number of words quite universally follow the family of Weibull distributions known from survival analyses. However, the values of two parameters marking specific forms of these distributions distinguish specific languages. This is such a strong constraint that the punctuation distributions of texts translated from the original language into another adopt quantitative characteristics of the target language. All these changes take place within Weibull distributions such that the corresponding hazard functions are always increasing. Recent previous research shows that James Joyce's famous "Finnegans Wake" is subject to such extreme distribution from the Weibull family that the corresponding hazard function is clearly decreasing. At the same time, the distances of sentence ending punctuation marks, determining the variability of sentence length, have an almost perfect multifractal organization, so far to such an extent found nowhere else in the literature. In the present contribution based on several available translations (Dutch, French, German, Polish, Russian) of "Finnegans Wake", it is shown that the punctuation characteristics of this work remain largely translation invariant, contrary to the common cases. These observations may constitute further evidence that "Finnegans Wake" is a translinguistic work in this respect as well, in line with Joyce's original intention.

As the recent studies indicate, the structure imposed onto written texts by the presence of punctuation develops patterns which reveal certain characteristics of universality. In particular, based on a large collection of classic literary works, it has been evidenced that the distances between consecutive punctuation marks, measured in terms of the number of words, obey the discrete Weibull distribution - a discrete variant of a distribution often used in survival analysis. The present work extends the analysis of punctuation usage patterns to more experimental pieces of world literature. It turns out that the compliance of the the distances between punctuation marks with the discrete Weibull distribution typically applies here as well. However, some of the works by James Joyce are distinct in this regard - in the sense that the tails of the relevant distributions are significantly thicker and, consequently, the corresponding hazard functions are decreasing functions not observed in typical literary texts in prose. "Finnegans Wake" - the same one to which science owes the word "quarks" for the most fundamental constituents of matter - is particularly striking in this context. At the same time, in all the studied texts, the sentence lengths - representing the distances between sentence-ending punctuation marks - reveal more freedom and are not constrained by the discrete Weibull distribution. This freedom in some cases translates into long-range nonlinear correlations, which manifest themselves in multifractality. Again, a text particularly spectacular in terms of multifractality is "Finnegans Wake".

Below is the text of the first page(about the same amount of characters as the first page) of the beloved by some, having bedeviled many, Finnegans Wake by James Joyce. But this time it's been fed through a neural network of machine learning(an RNN, a Recurrent Neural Network) meant to generate text off of a base text(the wonderful textgenrnn), along with its own byproduct of neuralfakewake. This was a quick project, I hesitate to even post the few lines of code it took, but credit to James Joyce(RIP, HCE), Max Woolf for the fantastically usable textgenrnn, and abstraction itself.

You can find some background for this post here: Deep Learning with Python! Torch has its strengths and its weaknesses. Caffe is much more popular that Torch, but when talking to some power users of Deep Learning (like Andrej Karpathy and other DeepMind scientists), a certain group of experts seems to be migrating from Caffe to Torch. I read somewhere else that Caffe: Torch:: Applications: Research. If you want to do serious research in deep learning, I would suggest using Torch given the level of current interest in the ecosystem, as well as Torch's flexibility and platform.