arXiv subjects

Sean Anderson

Publications and source records attributed to Sean Anderson.

12 recordsLinked to original sources

arXiv2026

Layer-wise Derivative Controlled Networks

As machine learning models grow in complexity, they increasingly struggle with three conflicting demands: the need for high accuracy, the requirement for hardware efficiency, and the necessity of functional stability. Traditional architectures often achieve performance at the expense of spiky or unpredictable behavior, where small changes in input lead to massive swings in output -- a critical flaw for real-world deployment in sensitive environments. This paper introduces ChainzRule (CR), a novel neural architecture designed to harmonize these competing goals. ChainzRule replaces standard piecewise-linear activations with a Polynomial Engine governed by Differential Regularization (DREG). Unlike traditional methods that impose global, coarse-grained constraints on a model's Lipschitz constant, DREG acts as a targeted regularization on intermediate derivatives. This approach suppresses extreme sensitivity without attenuating the representational power inherent in the Polynomial Engine. In head-to-head "Fair Fight" benchmarks, ChainzRule outperformed standard models while using 15.5x fewer parameters. On the MNIST dataset, it reduced peak gradient volatility by an average of 23.1%, ensuring a smoother and more predictable manifold. On Yelp Full ordinal regression under explicit DREG regularization, ChainzRule achieves 70.17% accuracy, validating that derivative-aware regularization is compatible with competitive performance on realistic tasks. By embedding gradient awareness into the architecture via DREG, ChainzRule demonstrates that stability and accuracy need not be competing objectives.

cs.LG↗

arXiv2025

Explainable Deep Convolutional Multi-Type Anomaly Detection

Explainable anomaly detection methods often have the capability to identify and spatially localise anomalies within an image but lack the capability to differentiate the type of anomaly. Furthermore, they often require the costly training and maintenance of separate models for each object category. The lack of specificity is a significant research gap because identifying the type of anomaly (e.g., "Crack" vs. "Scratch") is crucial for accurate diagnosis that facilitates cost-saving operational decisions across diverse application domains. While some recent large-scale Vision-Language Models (VLMs) have begun to address this, they are computationally intensive and memory-heavy, restricting their use in real-time or embedded systems. We propose MultiTypeFCDD, a simple and lightweight convolutional framework designed as a practical alternative for explainable multi-type anomaly detection. MultiTypeFCDD uses only image-level labels to learn and produce multi-channel heatmaps, where each channel is trained to correspond to a specific anomaly type. The model functions as a single, unified framework capable of differentiating anomaly types across multiple object categories, eliminating the need to train and manage separate models for each object category. We evaluated our proposed method on the Real-IAD dataset and it delivers competitive results (96.4% I-AUROC) at just over 1% the size of state-of-the-art VLM models used for similar tasks. This makes it a highly practical and viable solution for real-world applications where computational resources are tightly constrained.

cs.CV↗

arXiv2025

Zero-sum turn games using Q-learning: finite computation with security guarantees

This paper addresses zero-sum ``turn'' games, in which only one player can make decisions at each state. We show that pure saddle-point state-feedback policies for turn games can be constructed from dynamic programming fixed-point equations for a single value function or Q-function. These fixed-points can be constructed using a suitable form of Q-learning. For discounted costs, convergence of this form of Q-learning can be established using classical techniques. For undiscounted costs, we provide a convergence result that applies to finite-time deterministic games, which we use to illustrate our results. For complex games, the Q-learning iteration must be terminated before exploring the full-state, which can lead to policies that cannot guarantee the security levels implied by the final Q-function. To mitigate this, we propose an ``opponent-informed'' exploration policy for selecting the Q-learning samples. This form of exploration can guarantee that the final Q-function provides security levels that hold, at least, against a given set of policies. A numerical demonstration for a multi-agent game, Atlatl, indicates the effectiveness of these methods.

eess.SY↗

arXiv2025

Experiment Design with Gaussian Process Regression with Applications to Chance-Constrained Control

Learning for control in repeated tasks allows for well-designed experiments to gather the most useful data. We consider the setting in which we use a data-driven controller that does not have access to the true system dynamics. Rather, the controller uses inferred dynamics based on the available information. In order to acquire data that is beneficial for this controller, we present an experimental design approach that leverages the current data to improve expected control performance. We focus on the setting in which inference on the unknown dynamics is performed using Gaussian processes. Gaussian processes not only provide uncertainty quantification but also allow us to leverage structures inherent to Gaussian random variables. Through this structure, we design experiments via gradient descent on the expected control performance with respect to the experiment input. In particular, we focus on a chance-constrained minimum expected time control problem. Numerical demonstrations of our approach indicate our experimental design outperforms relevant benchmarks.

eess.SY↗

arXiv2024

Magnetic properties of diluted hexaferrites

We revisit the magnetic properties of the hexagonal ferrite PbFe$_{12-x}$Ga$_x$O$_{19}$. Recent experiments have reported puzzling dependencies of the ordering temperature and the saturation magnetization on the Ga concentration $x$. To explain these observations, we perform large-scale Monte Carlo simulations, focusing on the effects of an unequal distribution of the Ga impurities over the five distinct Fe sublattices. Ab-initio density-functional calculations predict that the Ga ions preferably occupy the $12k$ sublattice and (to a lesser extent) the $2a$ sublattice. We incorporate this insight into a nonuniform model of the Ga distribution. Monte Carlo simulations using this model lead to an excellent agreement between the theoretical and experimental values of the ordering temperature and saturation magnetization, indicating that the unequal distribution of the Ga impurities is the main reason for the unusual magnetic properties of PbFe$_{12-x}$Ga$_x$O$_{19}$. We also compute the temperature and concentration dependencies of the sublattice magnetizations, and we study the character of the zero-temperature transition that takes place when the ordering temperature is tuned to zero.

cond-mat.dis-nn↗

arXiv2024

Control-Oriented Identification for the Linear Quadratic Regulator: Technical Report

Data-driven control benefits from rich datasets, but constructing such datasets becomes challenging when gathering data is limited. We consider an offline experiment design approach to gathering data where we design a control input to collect data that will most improve the performance of a feedback controller. We show how such a control-oriented approach can be used in a setting with linear dynamics and quadratic objective and, through design of a gradient estimator, solve the problem via stochastic gradient descent. We show our formulation numerically outperforms an A- and L-optimal experiment design approach as well as a robust dual control approach.

eess.SY↗

arXiv2021

Explainable AI for medical imaging: Explaining pneumothorax diagnoses with Bayesian Teaching

Limited expert time is a key bottleneck in medical imaging. Due to advances in image classification, AI can now serve as decision-support for medical experts, with the potential for great gains in radiologist productivity and, by extension, public health. However, these gains are contingent on building and maintaining experts' trust in the AI agents. Explainable AI may build such trust by helping medical experts to understand the AI decision processes behind diagnostic judgements. Here we introduce and evaluate explanations based on Bayesian Teaching, a formal account of explanation rooted in the cognitive science of human learning. We find that medical experts exposed to explanations generated by Bayesian Teaching successfully predict the AI's diagnostic decisions and are more likely to certify the AI for cases when the AI is correct than when it is wrong, indicating appropriate trust. These results show that Explainable AI can be used to support human-AI collaboration in medical imaging.

cs.LG↗

arXiv2020

Towards Formally Verified Compilation of Tag-Based Policy Enforcement

Hardware-assisted reference monitoring is receiving increasing attention as a way to improve the security of existing software. One example is the PIPE architecture extension, which attaches metadata tags to register and memory values and executes tag-based rules at each machine instruction to enforce a software-defined security policy. To use PIPE effectively, engineers should be able to write security policies in terms of source-level concepts like functions, local variables, and structured control operators, which are not visible at machine level. It is the job of the compiler to generate PIPE-aware machine code that enforces these source-level policies. The compiler thus becomes part of the monitored system's trusted computing base -- and hence a prime candidate for verification. To formalize compiler correctness in this setting, we extend the source language semantics with its own form of user-specified tag-based monitoring, and show that the compiler preserves that monitoring behavior. The challenges of compilation include mapping source-level monitoring policies to instruction-level tag rules, preserving fail-stop behaviors, and satisfying the surprisingly complex preconditions for conventional optimizations. In this paper, we describe the design and verification of Tagine, a small prototype compiler that translates a simple tagged WHILE language to a tagged register transfer language and performs simple optimizations. Tagine is based on the RTLgen and Deadcode phases of the CompCert compiler, and hence is written and verified in Coq. This work is a first step toward verification of a full-scale compiler for a realistic tagged source language.

cs.CR↗

arXiv2019

Set-Based Adaptive Safety Control

Feedback Control Systems, ME C134/EE C128, is an introductory control systems course at UC Berkeley. Over the entire course, students gain practical experience by implementing various control schemes and designing observers in an effort to ultimately stabilize an inverted pendulum on a linear track. Throughout this learning process, frequent mishaps occur where improper controller implementation damages hardware. A simple example concerns the student's controller driving the cart into the wall at full speed. To offset the financial burden placed on the university in light of these mishaps, we designed a streamlined adaptive control system using set theory. We utilized lab-provided plant models to generate an $O_\infty$ set, attenuated the vertices to generate a safe, sub-region $S_\infty$, and attenuated in such a manner as to ensure an evolution of the vertices of $S_\infty$ remained within $O_\infty$ for at least one time step. Afterwards, we constructed a single Simulink block for students to easily implement within their own control schemes. This block consistently checks to see whether the system state remains within $S_\infty$. If that check is true, our controller does nothing. If it returns false, our controller takes over, drives the system to a prescribed safe-point, and shuts the system down. Overall, our process assumes perfect plant modelling, though our insistence on an evolution of $S_\infty$ remaining within $O_\infty$ resulted in considerable robustness to disturbances. In the end we were successful in implementing this real-time adaptive system and will provide it to the department for use in future labs.

eess.SY↗

arXiv2019

Electric Vehicle Charge Scheduling on Highway Networks from an Aggregate Cost Perspective

In this paper, we attempt to optimally schedule the charging of long-range battery electric vehicles (BEVs) along highway networks, in order to minimize aggregate costs to the overall system consisting of utilities or electricity providers, station operators and other infrastructure, as well as EV users. Thus, we approach the problem from the perspective of both customers (EV car owners), as well as charging station operators and utilities using a hybrid systems based formulation.

eess.SY↗

arXiv2017

Target-distractor Synchrony Affects Performance in a Novel Motor Task for Studying Action Selection

The study of action selection in humans can present challenges of task design since our actions are usually defined by many degrees of freedom and therefore occupy a large action-space. While saccadic eye-movement offers a more constrained paradigm for investigating action selection, the study of reach-and-grasp in upper limbs has often been defined by more complex scenarios, not easily interpretable in terms of such selection. Here we present a novel motor behaviour task which addresses this by limiting the action space to a single degree of freedom in which subjects have to track (using a stylus) a vertical coloured target line displayed on a tablet computer, whilst ignoring a similarly oriented distractor line in a different colour. We ran this task with 55 subjects and showed that, in agreement with previous studies, the presence of the distractor generally increases the movement latency and directional error rate. Further, we used two distractor conditions according to whether the distractor's location changes asynchronously or synchronously with the location of the target. We found that the asynchronous distractor yielded poorer performance than its synchronous counterpart, with significantly higher movement latencies and higher error rates. We interpret these results in an action selection framework with two actions (move left or right) and competing 'action requests' offered by the target and distractor. As such, the results provide insights into action selection performance in humans and supply data for directly constraining future computational models therein.

q-bio.NC↗

arXiv2014

Batch Nonlinear Continuous-Time Trajectory Estimation as Exactly Sparse Gaussian Process Regression

In this paper, we revisit batch state estimation through the lens of Gaussian process (GP) regression. We consider continuous-discrete estimation problems wherein a trajectory is viewed as a one-dimensional GP, with time as the independent variable. Our continuous-time prior can be defined by any nonlinear, time-varying stochastic differential equation driven by white noise; this allows the possibility of smoothing our trajectory estimates using a variety of vehicle dynamics models (e.g., `constant-velocity'). We show that this class of prior results in an inverse kernel matrix (i.e., covariance matrix between all pairs of measurement times) that is exactly sparse (block-tridiagonal) and that this can be exploited to carry out GP regression (and interpolation) very efficiently. When the prior is based on a linear, time-varying stochastic differential equation and the measurement model is also linear, this GP approach is equivalent to classical, discrete-time smoothing (at the measurement times); when a nonlinearity is present, we iterate over the whole trajectory to maximize accuracy. We test the approach experimentally on a simultaneous trajectory estimation and mapping problem using a mobile robot dataset.

cs.RO↗