Searcharxiv⌕ Search

arXiv subjects

Arend Hintze

Publications and source records attributed to Arend Hintze.

At least 19 recordsLinked to original sources
arXiv2025

Identifying Information-Transfer Nodes in a Recurrent Neural Network Reveals Dynamic Representations

Understanding the internal dynamics of Recurrent Neural Networks (RNNs) is crucial for advancing their interpretability and improving their design. This study introduces an innovative information-theoretic method to identify and analyze information-transfer nodes within RNNs, which we refer to as \textit{information relays}. By quantifying the mutual information between input and output vectors across nodes, our approach pinpoints critical pathways through which information flows during network operations. We apply this methodology to both synthetic and real-world time series classification tasks, employing various RNN architectures, including Long Short-Term Memory (LSTM) networks and Gated Recurrent Units (GRUs). Our results reveal distinct patterns of information relay across different architectures, offering insights into how information is processed and maintained over time. Additionally, we conduct node knockout experiments to assess the functional importance of identified nodes, significantly contributing to explainable artificial intelligence by elucidating how specific nodes influence overall network behavior. This study not only enhances our understanding of the complex mechanisms driving RNNs but also provides a valuable tool for designing more robust and interpretable neural networks.

cs.LG↗
arXiv2025

Functional Information Decomposition: A First-Principles Approach to Analyzing Functional Relationships

A central challenge in analyzing multivariate interactions within complex systems is to decompose how multiple inputs jointly determine an output. Existing approaches generally operate on observed probability distributions and can conflate a system's intrinsic functional logic with statistical artifacts of limited data. As a result, distinct systems can yield identical observations, rendering information decomposition fundamentally underdetermined and obscuring true higher-order interactions. We introduce Functional Information Decomposition (FID), both a computational and theoretical framework, which defines informational components with respect to a system's complete input-output mapping, thereby addressing a core cross-scale inference problem: determining how information carried by individual components combines to shape system-level behavior. When the mapping is fully specified, FID provides a unique decomposition into independent and synergistic contributions. Crucially, given only partial observations, FID characterizes the entire space of consistent decompositions by sampling compatible functions, making inferential limits explicit. A complementary geometric perspective clarifies the structural origin of informational components. We demonstrate FID's interdisciplinary utility on canonical logical functions, Conway's Game of Life, and gene-expression-based prediction of cancer drug response, and provide an open-source implementation. By separating functional architecture from observational distribution, FID offers a principled foundation for analyzing multivariate dependence in both fully and partially observed complex systems.

cs.IT↗
arXiv2025

Characterizing Fitness Landscape Structures in Prompt Engineering

While prompt engineering has emerged as a crucial technique for optimizing large language model performance, the underlying optimization landscape remains poorly understood. Current approaches treat prompt optimization as a black-box problem, applying sophisticated search algorithms without characterizing the landscape topology they navigate. We present a systematic analysis of fitness landscape structures in prompt engineering using autocorrelation analysis across semantic embedding spaces. Through experiments on error detection tasks with two distinct prompt generation strategies -- systematic enumeration (1,024 prompts) and novelty-driven diversification (1,000 prompts) -- we reveal fundamentally different landscape topologies. Systematic prompt generation yields smoothly decaying autocorrelation, while diversified generation exhibits non-monotonic patterns with peak correlation at intermediate semantic distances, indicating rugged, hierarchically structured landscapes. Task-specific analysis across 10 error detection categories reveals varying degrees of ruggedness across different error types. Our findings provide an empirical foundation for understanding the complexity of optimization in prompt engineering landscapes.

cs.AI↗
arXiv2025

Rethinking Self-Replication: Detecting Distributed Selfhood in the Outlier Cellular Automaton

Spontaneous self-replication in cellular automata has long been considered rare, with most known examples requiring careful design or artificial initialization. In this paper, we present formal, causal evidence that such replication can emerge unassisted -- and that it can do so in a distributed, multi-component form. Building on prior work identifying complex dynamics in the Outlier rule, we introduce a data-driven framework that reconstructs the full causal ancestry of patterns in a deterministic cellular automaton. This allows us to rigorously identify self-replicating structures via explicit causal lineages. Our results show definitively that self-replicators in the Outlier CA are not only spontaneous and robust, but are also often composed of multiple disjoint clusters working in coordination, raising questions about some conventional notions of individuality and replication in artificial life systems.

nlin.CG↗
arXiv2024

Promoting Cooperation in the Public Goods Game using Artificial Intelligent Agents

The tragedy of the commons illustrates a fundamental social dilemma where individual rational actions lead to collectively undesired outcomes, threatening the sustainability of shared resources. Strategies to escape this dilemma, however, are in short supply. In this study, we explore how artificial intelligence (AI) agents can be leveraged to enhance cooperation in public goods games, moving beyond traditional regulatory approaches to using AI as facilitators of cooperation. We investigate three scenarios: (1) Mandatory Cooperation Policy for AI Agents, where AI agents are institutionally mandated always to cooperate; (2) Player-Controlled Agent Cooperation Policy, where players evolve control over AI agents' likelihood to cooperate; and (3) Agents Mimic Players, where AI agents copy the behavior of players. Using a computational evolutionary model with a population of agents playing public goods games, we find that only when AI agents mimic player behavior does the critical synergy threshold for cooperation decrease, effectively resolving the dilemma. This suggests that we can leverage AI to promote collective well-being in societal dilemmas by designing AI agents to mimic human players.

cs.GT↗
arXiv2023

ChatGPT believes it is conscious

The development of advanced generative chat models, such as ChatGPT, has raised questions about the potential consciousness of these tools and the extent of their general artificial intelligence. ChatGPT consistent avoidance of passing the test is here overcome by asking ChatGPT to apply the Turing test to itself. This explores the possibility of the model recognizing its own sentience. In its own eyes, it passes this test. ChatGPT's self-assessment makes serious implications about our understanding of the Turing test and the nature of consciousness. This investigation concludes by considering the existence of distinct types of consciousness and the possibility that the Turing test is only effective when applied between consciousnesses of the same kind. This study also raises intriguing questions about the nature of AI consciousness and the validity of the Turing test as a means of verifying such consciousness.

cs.GL↗
arXiv2023

Detecting Information Relays in Deep Neural Networks

Deep learning of artificial neural networks (ANNs) is creating highly functional processes that are, unfortunately, nearly as hard to interpret as their biological counterparts. Identification of functional modules in natural brains plays an important role in cognitive and neuroscience alike, and can be carried out using a wide range of technologies such as fMRI, EEG/ERP, MEG, or calcium imaging. However, we do not have such robust methods at our disposal when it comes to understanding functional modules in artificial neural networks. Ideally, understanding which parts of an artificial neural network perform what function might help us to address a number of vexing problems in ANN research, such as catastrophic forgetting and overfitting. Furthermore, revealing a network's modularity could improve our trust in them by making these black boxes more transparent. Here, we introduce a new information-theoretic concept that proves useful in understanding and analyzing a network's functional modularity: the relay information $I_R$. The relay information measures how much information groups of neurons that participate in a particular function (modules) relay from inputs to outputs. Combined with a greedy search algorithm, relay information can be used to identify computational modules in neural networks. We also show that the functionality of modules correlates with the amount of relay information they carry.

cs.NE↗
arXiv2022

Incentivising cooperation by rewarding the weakest member

Autonomous agents that act with each other on behalf of humans are becoming more common in many social domains, such as customer service, transportation, and health care. In such social situations greedy strategies can reduce the positive outcome for all agents, such as leading to stop-and-go traffic on highways, or causing a denial of service on a communications channel. Instead, we desire autonomous decision-making for efficient performance while also considering equitability of the group to avoid these pitfalls. Unfortunately, in complex situations it is far easier to design machine learning objectives for selfish strategies than for equitable behaviors. Here we present a simple way to reward groups of agents in both evolution and reinforcement learning domains by the performance of their weakest member. We show how this yields ``fairer'' more equitable behavior, while also maximizing individual outcomes, and we show the relationship to biological selection mechanisms of group-level selection and inclusive fitness theory.

cs.MA↗
arXiv2020

Group-level selection avoids the tragedy of the commons

The public goods game is a famous example illustrating the tragedy of the commons. In this game cooperating individuals contribute to a pool, which in turn is distributed to all members of the group, including defectors who reap the same rewards as cooperators without having made a contribution before. The question is now, how to incentivize group members to all cooperate as it maximizes the common good. While costly punishment presents one such method, the cost of punishment still reduces the common good. Here we show how group-level selection can be such an incentive, and specifically how even fractions of group-level selection can overcome the benefits defectors receive. Further, we show how punishment and group-level selection interact. This work suggests that a redistribution similar to a basic income that is coupled to the economic success of the entire group could overcome the tragedy of the commons.

physics.soc-ph↗
arXiv2018

The structure of evolved representations across different substrates for artificial intelligence

Artificial neural networks (ANNs), while exceptionally useful for classification, are vulnerable to misdirection. Small amounts of noise can significantly affect their ability to correctly complete a task. Instead of generalizing concepts, ANNs seem to focus on surface statistical regularities in a given task. Here we compare how recurrent artificial neural networks, long short-term memory units, and Markov Brains sense and remember their environments. We show that information in Markov Brains is localized and sparsely distributed, while the other neural network substrates "smear" information about the environment across all nodes, which makes them vulnerable to noise.

cs.NE↗
arXiv2018

The Role of Conditional Independence in the Evolution of Intelligent Systems

Systems are typically made from simple components regardless of their complexity. While the function of each part is easily understood, higher order functions are emergent properties and are notoriously difficult to explain. In networked systems, both digital and biological, each component receives inputs, performs a simple computation, and creates an output. When these components have multiple outputs, we intuitively assume that the outputs are causally dependent on the inputs but are themselves independent of each other given the state of their shared input. However, this intuition can be violated for components with probabilistic logic, as these typically cannot be decomposed into separate logic gates with one output each. This violation of conditional independence on the past system state is equivalent to instantaneous interaction --- the idea is that some information between the outputs is not coming from the inputs and thus must have been created instantaneously. Here we compare evolved artificial neural systems with and without instantaneous interaction across several task environments. We show that systems without instantaneous interactions evolve faster, to higher final levels of performance, and require fewer logic components to create a densely connected cognitive machinery.

cs.AI↗
arXiv2017

Markov Brains: A Technical Introduction

Markov Brains are a class of evolvable artificial neural networks (ANN). They differ from conventional ANNs in many aspects, but the key difference is that instead of a layered architecture, with each node performing the same function, Markov Brains are networks built from individual computational components. These computational components interact with each other, receive inputs from sensors, and control motor outputs. The function of the computational components, their connections to each other, as well as connections to sensors and motors are all subject to evolutionary optimization. Here we describe in detail how a Markov Brain works, what techniques can be used to study them, and how they can be evolved.

cs.AI↗
arXiv2017

Thermodynamics of Evolutionary Games

How cooperation can evolve between players is an unsolved problem of biology. Here we use Hamiltonian dynamics of models of the Ising type to describe populations of cooperating and defecting players to show that the equilibrium fraction of cooperators is given by the expectation value of a thermal observable akin to a magnetization. We apply the formalism to the Public Goods game with three players, and show that a phase transition between cooperation and defection occurs that is equivalent to a transition in one-dimensional Ising crystals with long-range interactions. We then investigate the effect of punishment on cooperation and find that punishment plays the role of a magnetic field that leads to an "alignment" between players, thus encouraging cooperation. We suggest that a thermal Hamiltonian picture of the evolution of cooperation can generate other insights about the dynamics of evolving groups by mining the rich literature of critical dynamics in low-dimensional spin systems.

q-bio.PE↗
arXiv2017

Machine Learned Learning Machines

There are two common approaches for optimizing the performance of a machine: genetic algorithms and machine learning. A genetic algorithm is applied over many generations whereas machine learning works by applying feedback until the system meets a performance threshold. Though these are methods that typically operate separately, we combine evolutionary adaptation and machine learning into one approach. Our focus is on machines that can learn during their lifetime, but instead of equipping them with a machine learning algorithm we aim to let them evolve their ability to learn by themselves. We use evolvable networks of probabilistic and deterministic logic gates, known as Markov Brains, as our computational model organism. The ability of Markov Brains to learn is augmented by a novel adaptive component that can change its computational behavior based on feedback. We show that Markov Brains can indeed evolve to incorporate these feedback gates to improve their adaptability to variable environments. By combining these two methods, we now also implemented a computational model that can be used to study the evolution of learning.

cs.AI↗
arXiv2017

Origin of life in a digital microcosm

While all organisms on Earth descend from a common ancestor, there is no consensus on whether the origin of this ancestral self-replicator was a one-off event or whether it was only the final survivor of multiple origins. Here we use the digital evolution system Avida to study the origin of self-replicating computer programs. By using a computational system, we avoid many of the uncertainties inherent in any biochemical system of self-replicators (while running the risk of ignoring a fundamental aspect of biochemistry). We generated the exhaustive set of minimal-genome self-replicators and analyzed the network structure of this fitness landscape. We further examined the evolvability of these self-replicators and found that the evolvability of a self-replicator is dependent on its genomic architecture. We studied the differential ability of replicators to take over the population when competed against each other (akin to a primordial-soup model of biogenesis) and found that the probability of a self-replicator out-competing the others is not uniform. Instead, progenitor (most-recent common ancestor) genotypes are clustered in a small region of the replicator space. Our results demonstrate how computational systems can be used as test systems for hypotheses concerning the origin of life.

q-bio.PE↗
arXiv2016

Exploring the coevolution of predator and prey morphology and behavior

A common idiom in biology education states, "Eyes in the front, the animal hunts. Eyes on the side, the animal hides." In this paper, we explore one possible explanation for why predators tend to have forward-facing, high-acuity visual systems. We do so using an agent-based computational model of evolution, where predators and prey interact and adapt their behavior and morphology to one another over successive generations of evolution. In this model, we observe a coevolutionary cycle between prey swarming behavior and the predator's visual system, where the predator and prey continually adapt their visual system and behavior, respectively, over evolutionary time in reaction to one another due to the well-known "predator confusion effect." Furthermore, we provide evidence that the predator visual system is what drives this coevolutionary cycle, and suggest that the cycle could be closed if the predator evolves a hybrid visual system capable of narrow, high-acuity vision for tracking prey as well as broad, coarse vision for prey discovery. Thus, the conflicting demands imposed on a predator's visual system by the predator confusion effect could have led to the evolution of complex eyes in many predators.

q-bio.PE↗
arXiv2015

Evolvability tradeoffs in emergent digital replicators

The role of historical contingency in the origin of life is one of the great unknowns in modern science. Only one example of life exists--one that proceeded from a single self-replicating organism (or a set of replicating hyper-cycles) to the vast complexity we see today in Earth's biosphere. We know that emergent life has the potential to evolve great increases in complexity, but it is unknown if evolvability is automatic given any self-replicating organism. At the same time, it is difficult to test such questions in biochemical systems. Laboratory studies with RNA replicators have had some success with exploring the capacities of simple self-replicators, but these experiments are still limited in both capabilities and scope. Here, we use the digital evolution system Avida to explore the interaction between emergent replicators (rare randomly-assembled self-replicators) and evolvability. We find that we can classify fixed-length emergent replicators in Avida into two classes based on functional analysis. One class is more evolvable in the sense of optimizing their replication abilities. However, the other class is more evolvable in the sense of acquiring evolutionary innovations. We tie this trade-off in evolvability to the structure of the respective classes' replication machinery, and speculate on the relevance of these results to biochemical replicators.

q-bio.PE↗