arXiv subjects

Joseph West

Publications and source records attributed to Joseph West.

15 recordsLinked to original sources

arXiv2026

Visual Semantic Decoding of Electrocorticography from Video Stimuli using End-to-End Deep Learning

ECoG-based visual semantic decoding enables inference of semantic interpretation of visual perception from complex, noisy brain activity. This study examines the feasibility of visual semantic decoding using an end-to-end deep learning framework using electrocorticography (ECoG). Specifically, the decoding task is to predict visual categories from video stimuli using time-series neural inputs. A previously collected ECoG dataset from participants ($n=17$) with drug-resistant epilepsy is used for analysis. With fewer than 50 training samples per visual category, this study evaluates multiple deep learning approaches, artificial neural network architectures, and frequency-band filtered inputs. The best-performing approach is analyzed to shed light on the discriminative information it relies on across spectral, temporal, and cortical dimensions. The selected decoding system uses mixup augmentation, a Transformer-based encoder, and high-gamma (80-150 Hz) inputs with a 900 ms post-stimulus window. Further analysis shows that early visual cortex (V2-V4), ventral stream visual cortex, MT+ complex with neighbouring visual areas, and lateral temporal cortex contributed substantially to decoding performance. This study demonstrates that an end-to-end deep learning framework can yield promising decoding performance from dynamic visual stimuli without handcrafted features, while the model behavior remains interpretable through spectral, temporal, and cortical dimensions, which are broadly consistent with established neuroscience knowledge.

cs.LG↗

arXiv2026

Neural Distribution Prior for LiDAR Out-of-Distribution Detection

LiDAR-based perception is critical for autonomous driving due to its robustness to poor lighting and visibility conditions. Yet, current models operate under the closed-set assumption and often fail to recognize unexpected out-of-distribution (OOD) objects in the open world. Existing OOD scoring functions exhibit limited performance because they ignore the pronounced class imbalance inherent in LiDAR OOD detection and assume a uniform class distribution. To address this limitation, we propose the Neural Distribution Prior (NDP), a framework that models the distributional structure of network predictions and adaptively reweights OOD scores based on alignment with a learned distribution prior. NDP dynamically captures the logit distribution patterns of training data and corrects class-dependent confidence bias through an attention-based module. We further introduce a Perlin noise-based OOD synthesis strategy that generates diverse auxiliary OOD samples from input scans, enabling robust OOD training without external datasets. Extensive experiments on the SemanticKITTI and STU benchmarks demonstrate that NDP substantially improves OOD detection performance, achieving a point-level AP of 61.31% on the STU test set, which is more than 10$\times$ higher than the previous best result. Our framework is compatible with various existing OOD scoring formulations, providing an effective solution for open-world LiDAR perception.

cs.CV↗

arXiv2026

Encoding Matters: Benchmarking Binary and D-ary Representations for Quantum Combinatorial Optimization

Combinatorial optimization problems are typically formulated using Quadratic Unconstrained Binary Optimization (QUBO), where constraints are enforced through penalty terms that introduce auxiliary variables and rapidly increase Hamiltonian complexity, limiting scalability on near term quantum devices. In this work, we systematically study Quadratic Unconstrained D-ary Optimization (QUDO) as an alternative formulation in which decision variables are encoded directly in higher dimensional Hilbert spaces. We demonstrate that QUDO naturally captures structural constraints across a range of problem classes, including the Traveling Salesman Problem, two variants of the Vehicle Routing Problem, graph coloring, job scheduling, and Max-K-Cut, without the need for extensive penalty constructions. Using a qudit-level implementation of the Quantum Approximate Optimization Algorithm (qudit QAOA), we benchmark these formulations against their binary QUBO counterparts and exact classical solutions. Our study show consistently improved approximation ratios and substantially reduced computational overhead at comparable circuit depths, highlighting QUDO as a scalable and expressive representation for quantum combinatorial optimization.

quant-ph↗

arXiv2026

N-body Systems with Linear Wind Resistance: Analytic Solutions

The attractive and repulsive linear Hookean form of gravity is known to allow for analytic solutions to N-body systems for arbitrary masses and initial conditions. This linear system is very well suited for use in the advanced undergraduate classroom with the position of every particle rj(t) known analytically in terms of only sine, cosine, and exponential functions. Here it is shown that simple analytic solutions are also possible when linear wind resistance is included, provided that the force due to wind resistance force on each particle is proportional to the mass of that particle. Surprising analytic expressions for the behavior of the system if only a single particle in the system is subject to a linear wind resistance are also presented.

physics.class-ph↗

arXiv2025

Relative Energy Learning for LiDAR Out-of-Distribution Detection

Out-of-distribution (OOD) detection is a critical requirement for reliable autonomous driving, where safety depends on recognizing road obstacles and unexpected objects beyond the training distribution. Despite extensive research on OOD detection in 2D images, direct transfer to 3D LiDAR point clouds has been proven ineffective. Current LiDAR OOD methods struggle to distinguish rare anomalies from common classes, leading to high false-positive rates and overconfident errors in safety-critical settings. We propose Relative Energy Learning (REL), a simple yet effective framework for OOD detection in LiDAR point clouds. REL leverages the energy gap between positive (in-distribution) and negative logits as a relative scoring function, mitigating calibration issues in raw energy values and improving robustness across various scenes. To address the absence of OOD samples during training, we propose a lightweight data synthesis strategy called Point Raise, which perturbs existing point clouds to generate auxiliary anomalies without altering the inlier semantics. Evaluated on SemanticKITTI and the Spotting the Unexpected (STU) benchmark, REL consistently outperforms existing methods by a large margin. Our results highlight that modeling relative energy, combined with simple synthetic outliers, provides a principled and scalable solution for reliable OOD detection in open-world autonomous driving.

cs.CV↗

arXiv2025

Analytic Harmonic Tunable Closed Orbits of Trapped N-Body Systems

The motion of each particle in an N body system of identical masses interacting via an attractive or repulsive pairwise linear force law, the "Swarm," and with an external attractive or repulsive linear force law, the "Trap," is considered. In all Swarm and Trap combinations the motion of all N particles is completely separable and positions are found as a function of time in simple analytic form. in all attractive Traps the center of mass of the Swarm is bound to the center of the trap. For attractive or weakly repulsive Swarms in an attractive Trap the particles are bound to the center of mass of the Swarm and an infinite set of Swarm to Trap force constant ratios result in every particle executing a closed periodic orbit. Figures are provided of trajectories for various combinations of Swarm and Trap force constants. The derivations are suitable for the advanced undergraduate classroom.

physics.class-ph↗

arXiv2025

Adaptive Segmentation of EEG for Machine Learning Applications

Objective. Electroencephalography (EEG) data is derived by sampling continuous neurological time series signals. In order to prepare EEG signals for machine learning, the signal must be divided into manageable segments. The current naive approach uses arbitrary fixed time slices, which may have limited biological relevance because brain states are not confined to fixed intervals. We investigate whether adaptive segmentation methods are beneficial for machine learning EEG analysis. Approach. We introduce a novel adaptive segmentation method, CTXSEG, that creates variable-length segments based on statistical differences in the EEG data and propose ways to use them with modern machine learning approaches that typically require fixed-length input. We assess CTXSEG using controllable synthetic data generated by our novel signal generator CTXGEN. While our CTXSEG method has general utility, we validate it on a real-world use case by applying it to an EEG seizure detection problem. We compare the performance of CTXSEG with fixed-length segmentation in the preprocessing step of a typical EEG machine learning pipeline for seizure detection. Main results. We found that using CTXSEG to prepare EEG data improves seizure detection performance compared to fixed-length approaches when evaluated using a standardized framework, without modifying the machine learning method, and requires fewer segments. Significance. This work demonstrates that adaptive segmentation with CTXSEG can be readily applied to modern machine learning approaches, with potential to improve performance. It is a promising alternative to fixed-length segmentation for signal preprocessing and should be considered as part of the standard preprocessing repertoire in EEG machine learning applications.

cs.LG↗

arXiv2025

Out-of-distribution detection in 3D applications: a review

The ability to detect objects that are not prevalent in the training set is a critical capability in many 3D applications, including autonomous driving. Machine learning methods for object recognition often assume that all object categories encountered during inference belong to a closed set of classes present in the training data. This assumption limits generalization to the real world, as objects not seen during training may be misclassified or entirely ignored. As part of reliable AI, OOD detection identifies inputs that deviate significantly from the training distribution. This paper provides a comprehensive overview of OOD detection within the broader scope of trustworthy and uncertain AI. We begin with key use cases across diverse domains, introduce benchmark datasets spanning multiple modalities, and discuss evaluation metrics. Next, we present a comparative analysis of OOD detection methodologies, exploring model structures, uncertainty indicators, and distributional distance taxonomies, alongside uncertainty calibration techniques. Finally, we highlight promising research directions, including adversarially robust OOD detection and failure identification, particularly relevant to 3D applications. The paper offers both theoretical and practical insights into OOD detection, showcasing emerging research opportunities such as 3D vision integration. These insights help new researchers navigate the field more effectively, contributing to the development of reliable, safe, and robust AI systems.

cs.CV↗

arXiv2024

From Open Vocabulary to Open World: Teaching Vision Language Models to Detect Novel Objects

Traditional object detection methods operate under the closed-set assumption, where models can only detect a fixed number of objects predefined in the training set. Recent works on open vocabulary object detection (OVD) enable the detection of objects defined by an in-principle unbounded vocabulary, which reduces the cost of training models for specific tasks. However, OVD heavily relies on accurate prompts provided by an ``oracle'', which limits their use in critical applications such as driving scene perception. OVD models tend to misclassify near-out-of-distribution (NOOD) objects that have similar features to known classes, and ignore far-out-of-distribution (FOOD) objects. To address these limitations, we propose a framework that enables OVD models to operate in open world settings, by identifying and incrementally learning previously unseen objects. To detect FOOD objects, we propose Open World Embedding Learning (OWEL) and introduce the concept of Pseudo Unknown Embedding which infers the location of unknown classes in a continuous semantic space based on the information of known classes. We also propose Multi-Scale Contrastive Anchor Learning (MSCAL), which enables the identification of misclassified unknown objects by promoting the intra-class consistency of object embeddings at different scales. The proposed method achieves state-of-the-art performance on standard open world object detection and autonomous driving benchmarks while maintaining its open vocabulary object detection capability.

cs.CV↗

arXiv2024

N-body linear force law allowing analytic solutions

We present a pair-wise force law in a system of N particles that produces analytic solutions for arbitrary number of particles, masses, and initial conditions. Each pair of particles interacts via a force that is proportional to the product of their masses and their separation distance, with the force directed radially. We show that, despite the N-body interaction, each particle behaves as if it interacts only with the center of mass of the system. This effective two-body interaction behaves as Hooke's Law with a common frequency for all particles, with the familiar analytic solutions for the trajectories. With these analytic solutions, it is possible to efficiently simulate a collection of these particles and incorporate other external forces. As an example, we simulate the particles within an adiabatically expanding container and calculate pressure and temperature in both the attractive and repulsive cases.

physics.class-ph↗

arXiv2024

The ZK Spiral Family of Trajectories and Forces

Particle trajectories in the form of a logarithmic spiral with specified angular time dependence, "ZK spirals," are shown to be analytic solutions for motion in non-central, but simple force power-laws. Each ZK spiral is a particular solution to a single associated force law. The position and velocity are determined as analytic functions of time in terms of the growth parameter, specified power law, and the initial conditions. Four examples, each for a different force law and relevant to the advanced classroom, are presented: the well-known attractive central inverse cube force; a bead on a rigid, horizontal, frictionless wire in the shape of the spiral trajectory; a car moving in a changing radius turn; and a known solution for a powered rocket with variable thrust in a Newtonian gravitational field. In this last case, general expressions for position and velocity of powered transfer orbits within the solar system are presented along with a new analytic expression for the rocket mass as a function of time or distance from the Sun. These results can be extended to include powered flight solutions to a generalized Lambert's problem for circular orbits in attractive central force laws of the form F(r) = - Fo/(r^q), for q greater than or equal to 1.

physics.class-ph↗

arXiv2019

Improved Reinforcement Learning with Curriculum

Humans tend to learn complex abstract concepts faster if examples are presented in a structured manner. For instance, when learning how to play a board game, usually one of the first concepts learned is how the game ends, i.e. the actions that lead to a terminal state (win, lose or draw). The advantage of learning end-games first is that once the actions which lead to a terminal state are understood, it becomes possible to incrementally learn the consequences of actions that are further away from a terminal state - we call this an end-game-first curriculum. Currently the state-of-the-art machine learning player for general board games, AlphaZero by Google DeepMind, does not employ a structured training curriculum; instead learning from the entire game at all times. By employing an end-game-first training curriculum to train an AlphaZero inspired player, we empirically show that the rate of learning of an artificial player can be improved during the early stages of training when compared to a player not using a training curriculum.

cs.LG↗

arXiv2015

Building the next pyramid

The results of experimental tests of a novel method for moving large (pyramid construction size) stone blocks by rolling them are presented. The method is implemented by tying 12 identical rods of appropriately chosen radius to the faces of the block forming a rough dodecagon prism. Experiments using a 1,000 kg block show that it can be moved across level open ground with a dynamic coefficient of friction of less than 0.06. This value is a factor of five lower than that obtained for dragging the block, and the best values reported for dragging by others, at 0.3. the results are more dramatic than those obtained on smaller scale experiments on a 29.6 kg block, also reported here. For full scale pyramid blocks, the wooden "rods" woudl need to be posts of order 30 cm in diameter, similar in size to those used as masts on ships in the Nile.

physics.pop-ph↗

arXiv2012

Blackbody Friction: Analytic expressions for velocity and position

The equations of motion are solved analytically for speed and position for the case of blackbody friction on objects traveling at relativistic speeds with respect to observers fixed in the frame of the blackbody radiation. Two model cases are considered, both assume an object which is a perfect absorber of light, and one of which also assumes that the object maintains a constant rest mass by emitting photons in a momentum neutral process. The maximum distances attained for a light sail and a micron-size sand grain are determined under the assumption that the relevant blackbody source is that of cosmic background radiation. At a temperature of 3000K, the temperature at the time the universe became transparent, cosmic background radiation would cause a significant decay of speed fluctuations in dust clouds.

physics.class-ph↗

arXiv2010

Rolling Release Siege Engines: Teaching an Old Machine a New Trick

The analysis of a new "rolling release" mechanism is presented for three different "siege engine" designs. The range obtained with the rolling release is compared to the ranges obtained using the simpler "cup" and the proven "sling" release mechanisms. It is found that the rolling release is a significant improvement over the cup release, but the range of the rolling release is still well short of that attained with the sling release. It is also noted that the familiar "spoon" projectile holder used on many siege engines operates in a rolling release manner, contrary to appearances and to some previous published studies. Instead, the release mechanism used in ancient engines and in modern kit-purchased replicas is the rolling release.

physics.pop-ph↗