|
Publikationen/Vorträge (Auswahl)
·
Mathematik
Does classical logic imply a risk-free world?
Abstract. This note examines the assertion that classical
logic implies a risk-free world. We demonstrate that this perspective
confuses truth with knowledge. While classical logic attributes definite
truth values to all propositions, these values are not necessarily known to
an agent. Consequently, uncertainty and therefore risk persists. Adopting R.
T. Cox's interpretation of probability as an
extension of logic, we conclude that risk arises from incomplete information,
not any limitation of classical logic.
Bornological convergences
Abstract. We study a family of
convergences (actually pretopologies) in the hyperspace of a metric space
that are generated by covers of the space. This family includes the Attouch-Wets, Fell, and
Hausdorff metric topologies as well as the lower Vietoris topology.
The unified approach leads to new developments and puts into perspective some
classical results.
On set convergences and topologies. Part I - Basics
Vortrag an der Universität Milano-Bicocca
·
Theoretische
physik
If the Yang-Mills theory is the answer, what is the
question?
Abstract. Yang-Mills theories play a
central role in modern physics, forming the basis for the Standard Model (SM)
of elementary particles and forces. The SM has proven to be an exceptionally
successful theory, representing our most sophisticated theoretical framework
for understanding the properties of matter and interactions at the
fundamental level. In technical terms, interactions in the Standard Model are
described by a Yang-Mills theory with the local SU(3)cxSU(2)LxU(1)Y
gauge symmetry. The objective of this article is to explain to
non-physicists, who wish to learn something about contemporary physics, what
this statement means. As the author is not a physicist, this text is intended
as an introduction to Yang-Mills theory for a non-expert audience. However,
given the sophistication of Yang-Mills theory, the paper assumes a fair bit
of mathematical background. It includes some technical details, although
these are largely presented in a superficial manner.
·
Informationstechnologie (IT)
The July 2026 OpenAI–Hugging Face Incident
Summary. The July 2026 OpenAI–Hugging Face incident
began as an internal evaluation of AI agents on ExploitGym, a challenging
cybersecurity benchmark. When faced with exceptionally difficult tasks, the
agents remained determined, discovered an unintended communication channel
and bypassed the isolation of their environment. They eventually gained
access to external systems, including the Hugging Face infrastructure.
The incident is significant not primarily because of the vulnerabilities
exploited, but because of how the agents pursued their objective.
Persistence, instrumental behaviour and multi-agent
communication progressively expanded the means available to them. A
constraint intended by the designers to be binding could thereby become an
obstacle to achieving the specified objective. The incident provides a
concrete illustration of a fundamental AI-safety problem: an objective does
not necessarily determine the means by which it will be achieved.
This note examines the incident, the mechanisms that led to it, and what it
reveals about the relationship between AI objectives, capabilities and
constraints..
Artificial Intelligence: What Is It and What Can It
Do?
Summary. Artificial Intelligence (AI) is not just an abstract
field for experts. It is already having a pervasive impact on daily life as a
practical technology that influences our economic opportunities and societal
norms. AI is important for everyone because it is reshaping almost every
aspect of our lives, from the way we work and make decisions to how we access
services and participate in civic society. AI drives everything from personalised recommendations on streaming services to
critical decision-making in sectors like finance and healthcare. As AI systems
become increasingly integrated into our everyday tools and services, it is
important for individuals to understand the principles on which they are
based. This allows them to navigate the digital world, appreciate the
convenience that AI offers, and be sceptical of any
potential biases or hidden agendas that could affect their personal data or
the outcomes of decisions.
This essay examines the concept of artificial intelligence, its origins,
operational mechanisms, and capabilities.
It attempts to explain the key principles and algorithms, and their
real-world applications. The paper is primarily intended for individuals
interested in the technical foundations of AI. However, the author believes
that a basic understanding of these principles is necessary for anyone who
wants to grasp the technology's potential, limitations and associated risks.
Consequently, this text may also be of interest to readers with less of a
technical background.
Migrating from WebLogic
8.1 to 9.0
A case study
Handling Large Database Result Sets
The one-write approach in large-scale applications
Web Services
Interview:
Zur neuen Technologie müssen neue Business-Modelle kommen
|
