arXiv subjects

Mark Fisher

Publications and source records attributed to Mark Fisher.

4 recordsLinked to original sources

arXiv2025

Making AI Inevitable: Historical Perspective and the Problems of Predicting Long-Term Technological Change

This study demonstrates the extent to which prominent debates about the future of AI are best understood as subjective, philosophical disagreements over the history and future of technological change rather than as objective, material disagreements over the technologies themselves. It focuses on the deep disagreements over whether artificial general intelligence (AGI) will prove transformative for human society; a question that is analytically prior to that of whether this transformative effect will help or harm humanity. The study begins by distinguishing two fundamental camps in this debate. The first of these can be identified as "transformationalists," who argue that continued AI development will inevitably have a profound effect on society. Opposed to them are "skeptics," a more eclectic group united by their disbelief that AI can or will live up to such high expectations. Each camp admits further "strong" and "weak" variants depending on their tolerance for epistemic risk. These stylized contrasts help to identify a set of fundamental questions that shape the camps' respective interpretations of the future of AI. Three questions in particular are focused on: the possibility of non-biological intelligence, the appropriate time frame of technological predictions, and the assumed trajectory of technological development. In highlighting these specific points of non-technical disagreement, this study demonstrates the wide range of different arguments used to justify either the transformationalist or skeptical position. At the same time, it highlights the strong argumentative burden of the transformationalist position, the way that belief in this position creates competitive pressures to achieve first-mover advantage, and the need to widen the concept of "expertise" in debates surrounding the future development of AI.

cs.CY↗

arXiv2018

Self-ensembling for visual domain adaptation

This paper explores the use of self-ensembling for visual domain adaptation problems. Our technique is derived from the mean teacher variant (Tarvainen et al., 2017) of temporal ensembling (Laine et al;, 2017), a technique that achieved state of the art results in the area of semi-supervised learning. We introduce a number of modifications to their approach for challenging domain adaptation scenarios and evaluate its effectiveness. Our approach achieves state of the art results in a variety of benchmarks, including our winning entry in the VISDA-2017 visual domain adaptation challenge. In small image benchmarks, our algorithm not only outperforms prior art, but can also achieve accuracy that is close to that of a classifier trained in a supervised fashion.

cs.CV↗

arXiv2011

There are no magnetically charged particle-like solutions of the Einstein Yang-Mills equations for Abelian models

We prove that there are no magnetically charged particle-like solutions for Abelian models in Einstein Yang-Mills, but for non-Abelian models the possibility remains open. An analysis of the Lie algebraic structure of the Yang-Mills fields is essential to our results. In one key step of our analysis we use invariant polynomials to determine which orbits of the gauge group contain the possible asymptotic Yang-Mills field configurations. Together with a new horizontal/vertical space decomposition of the Yang-Mills fields this enables us to overcome some obstacles and complete a dynamical system existence theorem for asymptotic solutions with nonzero total magnetic charge. We then prove that these solutions cannot be extended globally for Abelian models and begin an investigation of the details for non-Abelian models.

gr-qc↗

arXiv2009

Static Spherically Symmetric Solutions of the SO(5) Einstein Yang-Mills Equations

Globally regular (ie. asymptotically flat and regular interior), spherically symmetric and localised ("particle-like") solutions of the coupled Einstein Yang-Mills (EYM) equations with gauge group SU(2) have been known for more than 20 years, yet their properties are still not well understood. Spherically symmetric Yang--Mills fields are classified by a choice of isotropy generator and SO(5) is distinguished as the simplest model with a \emph{non-Abelian} residual (little) group, $SU(2)\times U(1)$, and which admits globally regular particle-like solutions. We exhibit an algebraic gauge condition which normalises the residual gauge freedom to a finite number of discrete symmetries. This generalises the well-known reduction to the real magnetic potential $w(r,t)$ in the original SU(2) YM model. Reformulating using gauge invariant polynomials dramatically simplifies the system and makes numerical search techniques feasible. We find three families of embedded SU(2) EYM equations within the SO(5) system, one of which was first detected only within the gauge-invariant polynomial reduced system. Numerical solutions representing mixtures of the three SU(2) sub-systems are found, classified by a pair of positive integers.

gr-qc↗