arXiv subjects

Thomas Chaffey

Publications and source records attributed to Thomas Chaffey.

At least 19 recordsLinked to original sources

arXiv2026

Who Is Doing the Thinking? Characterising Human Agency in First-Year Students' AI Interactions

When students use generative AI to produce their work, the learning process becomes invisible. Established frameworks such as Bloom's taxonomy then classify the cognitive level of what the AI produced, not what the student contributed. We propose the Human Agency Depth Taxonomy (HADT), a preliminary six-tier taxonomy that classifies the engineering contribution a student makes in a prompt, independent of the AI's output. HADT combines three dimensions: contextual bandwidth, engineering abstraction and student agency. We applied HADT to 962 prompts from 105 students' structured AI Journals in an AI-permissive, first-year software engineering course of 145 students, and examined how agency changed over ten journalled weeks. Most prompts (84.2%) carried some student contribution beyond delegation. Agency tracked the development lifecycle. Architecture-discovery prompts dominated the requirements phase (48% of turns) and fell to 5-12% from week 9, while prompts analysing existing systems rose during build and verification (odds ratio 2.88, with turns clustered by student). Two-thirds of the variance in tier lay within students, so agency depended more on the phase of work than on the student. A turn-level rise in delegation in deadline weeks did not hold once turns were clustered by student. Explicitly specifying constraints on the AI's output was rare throughout, which we interpret as a knowledge bottleneck for novices rather than a gap in instruction. We discuss implications for course and assessment design, and for using HADT as a formative lens on how students engage with AI.

cs.SE↗

arXiv2026

Splitting Compositions of Monotone Operators

Coupled convex optimisation equilibria, bilinear saddle and potential systems, and nonlinear electronic device laws lead naturally to inclusions of the form $0\in Hx+A(Bx)-b$, where $A$ and $B$ are maximal monotone but their composition need not be monotone. We introduce a new class of splitting algorithms that solve such inclusions. Introducing the intermediate graph variable $y\in Bx$ exposes a directed two-row system. A one-parameter family of kernels produces a lower-triangular warped-resolvent iteration evaluated with one application of $H$ and the ordinary resolvents $J_{αA}$ and $J_{βB}$. Appropriate specializations of the problem data and algorithmic parameters recover forward--backward, inertial, primal--dual, and Douglas--Rachford constructions, while setting $H=b=0$ generates three novel algorithms for the pure composition $0\in A(Bx)$. Convergence of the coupled method is certified by an explicit nonnegative two-block gain matrix. Its Schur condition yields contraction in particular weighted norms, global linear convergence, and residual error estimates. For the general problem class, our convergence condition is shown to be sharp for an explicitly constructed linear family. When the problem is specialized to a pure composition, we prove convergence under the weaker assumption that only one operator is strongly monotone. This weaker regime is illustrated on a nonlinear circuit problem. We further illustrate our methods by solving inference problems for cascaded equilibrium neural networks.

math.OC↗

arXiv2026

Beyond Correctness: How Novice Software Engineering Students Exercise Oversight of Generative AI

Generative AI (GenAI) is shifting software engineering students' work from producing artefacts towards evaluating and supervising AI-generated outputs. However, little is known about what characteristics students identify in outputs when reporting interventions, or how they exercise oversight during authentic software engineering work. We analysed 962 student-AI interaction turns documented by 145 first-year software engineering students through structured AI Journals, including 172 student-declared interventions. Using a multi-dimensional coding framework, we characterised the output characteristics identified in these reported interventions, students' reasons for intervening, the actions they took, and how those actions were enacted. Fitness problems were identified in 74.4% of interventions, compared with 16.9% involving correctness problems, indicating that oversight more often concerned whether plausible AI outputs were suitable for the engineering task than whether they contained explicit errors. Among classifiable interventions, students most frequently modified generated outputs (51.0%) or refined their prompts (28.2%), with both response types commonly associated with engineering fit. Within-student analysis further showed that outputs perceived as containing minor flaws were associated with lower evaluation confidence than outputs perceived as perfect. Taken together, these findings characterise novice AI oversight as a process that extends beyond error detection to encompass both judgement of fitness for purpose and active intervention in AI-assisted engineering work.

cs.SE↗

arXiv2026

The Reciprocal tau-Cell: A Log-Domain Circuit for On-Chip Learning

Training hardware by gradient descent requires a gradient to be computed using a separate backwards pass. If a circuit is reciprocal, meaning every connection is symmetric, the backward pass is the same network run a second time with the error as its input. The reciprocal tau-cell, a weak-inversion log-domain circuit, makes this practical in standard CMOS. Each cell is a current-mode leaky integrator with a built-in ReLU; each connection is two transistors whose signed strength is a ratio of bias currents. The equilibrium of the network is the solution to a convex program, and latching the cells that are off will hardware linearise the circuit, making its second run exactly linear. Transistor-level SKY130 simulations show that the second run gives the gradient within 1.9%; that hardware linearisation permits large, noise-robust error signals; that a 13-cell network learns the Iris data set; and that learning in situ restores mismatched circuits on which software-trained weights fail, from 53-63% to 88-96% test accuracy.

eess.SP↗

arXiv2026

Amplitude-Dependent Bode Diagrams via Scaled Relative Graphs

Scaled Relative Graphs (SRGs) provide an intuitive graphical frequency-domain method for the analysis of Nonlinear (NL) systems, generalizing the Nyquist diagram. In this paper, we develop a method for computing $L_2$-gain bounds for Lur'e systems over bounded frequency and amplitude ranges. We do this by restricting the input space of the SRG both in frequency and energy content, and using methods from Sobolev theory. The resulting gain bounds over restricted sets of inputs are less conservative than bounds computed over the entire $L_2$, and yield three-dimensional NL generalization of the Bode diagram, plotting $L_2$-gain as function of both input frequency and energy content. In the zero-energy limit, the Linear Time-Invariant (LTI) Bode diagram is recovered, and at the infinite-energy zero-frequency limit, we recover the $L_2$-gain. The effectiveness of our method is demonstrated on an example that resembles Phase-Locked Loop dynamics.

eess.SY↗

arXiv2026

Circuit realization and hardware linearization of monotone operator equilibrium networks

It is shown that the port behavior of a resistor-diode network corresponds to the solution of a ReLU monotone operator equilibrium network (a neural network in the limit of infinite depth), giving a parsimonious construction of a neural network in analog hardware. We furthermore show that the gradient of such a circuit can be computed directly in hardware, using a procedure we call hardware linearization. This allows the network to be trained in hardware, which we demonstrate with a device-level circuit simulation. We extend the results to cascades of resistor-diode networks, which can be used to implement feedforward and other asymmetric networks. We finally show that different nonlinear elements give rise to different activation functions, and introduce the novel diode ReLU which is induced by a non-ideal diode model.

eess.SY↗

arXiv2026

Making the Invisible Visible: A Framework for Reflective AI Use in Software Engineering Education

Generative AI (GenAI) is increasingly embedded in software engineering education, supporting activities such as requirements development, design exploration, documentation, and prototyping. However, educators often have visibility only into final artefacts, with limited insight into how students evaluate, verify, and refine AI-generated outputs during the learning process. This creates challenges for assessing evaluative judgement and responsible AI-assisted practice. This paper introduces the AI Journal, a structured reflection framework designed to make student-GenAI interaction visible in first-year software engineering education. The framework combines execution tracking, which records prompts, outputs, intent, and interaction context, with cognitive auditing, which captures verification strategies, intervention decisions, confidence judgements, critical learning moments, and reflections on AI-supported work. Deployed in a first-semester software engineering course, the AI Journal enabled visibility into aspects of student learning not observable through artefact-based assessments alone. Preliminary observations suggested variation in verification practices, intervention strategies, and perceptions of AI-supported work. Critical learning moments frequently occurred when students evaluated contextual suitability, feasibility, and requirements alignment rather than identifying obvious errors. The AI Journal demonstrates a practical, lightweight, and model-agnostic approach for making AI-assisted learning processes visible. By foregrounding verification, intervention, and reflection, it shifts attention from product-focused assessment toward evaluative judgement and responsible AI-assisted practice.

cs.SE↗

arXiv2026

Quantization Robustness of Monotone Operator Equilibrium Networks

Monotone operator equilibrium networks are implicit-layer models whose output is the unique equilibrium of a monotone operator, guaranteeing existence, uniqueness, and convergence. When deployed on low-precision hardware, weights are quantized, potentially destroying these guarantees. We analyze weight quantization as a spectral perturbation of the underlying monotone inclusion. Convergence of the quantized solver is guaranteed whenever the spectral-norm weight perturbation is smaller than the monotonicity margin; the displacement between quantized and full-precision equilibria is bounded in terms of the perturbation size and margin; and a condition number characterizing the ratio of the operator norm to the margin links quantization precision to forward error. MNIST experiments confirm a phase transition at the predicted threshold: three- and four-bit post-training quantization diverge, while five-bit and above converge. The backward-pass guarantee enables quantization-aware training, which recovers provable convergence at four bits.

math.OC↗

arXiv2026

Convergence of energy-based learning in linear resistive networks

Energy-based learning algorithms are alternatives to backpropagation and are well-suited to distributed implementations in analog electronic devices. However, a rigorous theory of convergence is lacking. We make a first step in this direction by analysing a particular energybased learning algorithm, Contrastive Learning, applied to a network of linear adjustable resistors. It is shown that, in this setup, Contrastive Learning is equivalent to projected gradient descent on a convex function with Lipschitz continuous gradient, giving a guarantee of convergence of the algorithm for a range of stepsizes. This convergence result is then extended to a stochastic variant of Contrastive Learning.

math.OC↗

arXiv2025

Trust in LLM-controlled Robotics: a Survey of Security Threats, Defenses and Challenges

The integration of Large Language Models (LLMs) into robotics has revolutionized their ability to interpret complex human commands and execute sophisticated tasks. However, such paradigm shift introduces critical security vulnerabilities stemming from the ''embodiment gap'', a discord between the LLM's abstract reasoning and the physical, context-dependent nature of robotics. While security for text-based LLMs is an active area of research, existing solutions are often insufficient to address the unique threats for the embodied robotic agents, where malicious outputs manifest not merely as harmful text but as dangerous physical actions. In this work, we present a systematic survey, summarizing the emerging threat landscape and corresponding defense strategies for LLM-controlled robotics. Specifically, we discuss a comprehensive taxonomy of attack vectors, covering topics such as jailbreaking, backdoor attacks, and multi-modal prompt injection. In response, we analyze and categorize a range of defense mechanisms, from formal safety specifications and runtime enforcement to multi-LLM oversight and prompt hardening. Furthermore, we review key datasets and benchmarks used to evaluate the robustness of these embodied systems. By synthesizing current research, this work highlights the urgent need for context-aware security solutions and provides a foundational roadmap for the development of safe, secure, and reliable LLM-controlled robotics.

cs.RO↗

arXiv2025

Kernel Modelling of Fading Memory Systems

The paper is a follow-up of the recently introduced kernel-based framework to identify nonlinear input-output systems regularized by desirable input-output incremental properties. Assuming that the system has fading memory, we propose to learn the functional that maps the past input to the present output rather than the operator mapping input trajectories to output trajectories. While retaining the benefits of the previously proposed framework, this modification simplifies the selection of the kernel, enforces causality, and enables temporal simulation.

eess.SY↗

arXiv2025

Operator-Splitting Methods for Neuromorphic Circuit Simulation

A novel splitting algorithm is proposed for the numerical simulation of neuromorphic circuits. The algorithm is grounded in the operator-theoretic concept of monotonicity, which bears both physical and algorithmic significance. The splitting exploits this correspondence to translate the circuit architecture into the algorithmic architecture. The paper illustrates the many advantages of the proposed operator-theoretic framework over conventional numerical integration for the simulation of multiscale hierarchical events that characterize neuromorphic behaviors.

eess.SY↗

arXiv2025

An Operator-Theoretic Framework to Simulate Neuromorphic Circuits

Splitting algorithms are well-established in convex optimization and are designed to solve large-scale problems. Using such algorithms to simulate the behavior of nonlinear circuit networks provides scalable methods for the simulation and design of neuromorphic systems. For circuits made of linear capacitors and inductors with nonlinear resistive elements, we propose a splitting that breaks the network into its LTI lossless component and its static resistive component. This splitting has both physical and algorithmic advantages and allows for separate calculations in the time domain and in the frequency domain. To demonstrate the scalability of this approach, a network made from one hundred neurons modeled by the well-known FitzHugh-Nagumo circuit with all-to-all diffusive coupling is simulated.

eess.SY↗

arXiv2025

On phase in scaled graphs

The scaled graph has been introduced recently as a nonlinear extension of the classical Nyquist plot for linear time-invariant systems. In this paper, we introduce a modified definition for the scaled graph, termed the signed scaled graph (SSG), in which the phase component is characterized by making use of the Hilbert transform. Whereas the original definition of the scaled graph uses unsigned phase angles, the new definition has signed phase angles which ensures the possibility to differentiate between phase-lead and phase-lag properties in a system. Making such distinction is important from both an analysis and a synthesis perspective, and helps in providing tighter stability estimates of feedback interconnections. We show how the proposed SSG leads to intuitive characterizations of positive real and negative imaginary nonlinear systems, and present various interconnection results. We showcase the effectiveness of our results through several motivating examples.

eess.SY↗

arXiv2025

Graphical Dominance Analysis for Linear Systems: A Frequency-Domain Approach

We propose a frequency-domain approach to dominance analysis for multi-input multi-output (MIMO) linear time-invariant systems. The dominance of a MIMO system is defined to be the number of its poles in the open right half-plane. Our approach is graphical: we define a frequency-wise notion of the recently-introduced scaled graph of a MIMO system plotted in a complex plane. The scaled graph provides a bound of the eigenloci of the system, which can be viewed as a robust MIMO extension of the classical Nyquist plot. Our main results characterize sufficient conditions for quantifying the dominance of a closed-loop system based upon separation of scaled graphs of two open-loop systems in a frequency-wise manner. The results reconcile existing small gain, small phase and passivity theorems for feedback dominance analysis.

eess.SY↗

arXiv2025

Amplitude response and square wave describing functions

An analog of the describing function method is developed using square waves rather than sinusoids. Static nonlinearities map square waves to square waves, and their behavior is characterized by their response to square waves of varying amplitude - their amplitude response. The output of an LTI system to a square wave input is approximated by a square wave, to give an analog of the describing function. The classical describing function method for predicting oscillations in feedback interconnections is generalized to this square wave setting, and gives accurate predictions when oscillations are approximately square.

eess.SY↗

arXiv2024

A homotopy theorem for incremental stability

A theorem is proved to verify incremental stability of a feedback system via a homotopy from a known incrementally stable system. A first corollary of that result is that incremental stability may be verified by separation of Scaled Relative Graphs, correcting two assumptions in [1, Theorem 2]. A second corollary provides an incremental version of the classical IQC stability theorem.

math.OC↗

arXiv2024

Characterizing nonlinear systems with mixed input-output properties through dissipation inequalities

Systems that show different characteristics, such as finite-gain and passivity, depending on the nature of the inputs, are said to possess mixed input-output properties. In this paper, we provide a constructive method for characterizing mixed input-output properties of nonlinear systems using a dissipativity framework. Our results take inspiration from the generalized Kalman-Yakubovich-Popov lemma, and show that a system is mixed if it is dissipative with respect to highly specialized supply rates. The mixed input-output characterization is used for assessing stability of feedback interconnections in which the feedback components violate conditions of classical results such as the small-gain and passivity theorem. We showcase applicability of our results through various examples.

eess.SY↗