arXiv subjects

Michael Chin

Publications and source records attributed to Michael Chin.

6 recordsLinked to original sources

arXiv2026

Performance Verification of the AmpereOne CPU Core

As process technology scaling slows, microarchitectural innovation has become the primary driver of performance gains, making pre-silicon Performance Verification (PV) more critical than ever. This paper presents the industrial-scale PV methodology applied across four generations of the AmpereOne custom CPU core, centered on the cycle-accurate correlation of the RTL design against a trace-driven performance model. The methodology integrates data-driven workload curation, a high-frequency daily regression system, and a unified event-stream framework for analysis. We demonstrate this methodology through case studies of the Branch Prediction Unit and L2 Prefetcher, highlighting a hierarchical strategy that first isolates individual units for focused correlation before proceeding to full-core verification. The results demonstrate that this disciplined, iterative process is indispensable for avoiding costly post-silicon bugs and ensuring complex processors meet their performance targets. We end with a look towards the future of PV in the microprocessor industry.

cs.AR↗

arXiv2026

Conformal Prediction for Neural Operators: Distribution-Free Uncertainty Quantification in Physics Simulation

Neural operators such as the Fourier Neural Operator (FNO) have emerged as powerful surrogates for solving partial differential equations (PDEs), achieving speedups of several orders of magnitude over traditional numerical solvers. However, deploying these models in safety-critical engineering applications -- such as thermal management of electronic components and battery systems -- requires not only accurate point predictions but also rigorous uncertainty guarantees. Existing uncertainty quantification (UQ) methods for neural operators, including Monte Carlo Dropout and Deep Ensembles, provide only relative uncertainty estimates without formal coverage guarantees. In this work, we propose the first application of split conformal prediction to neural operator-based physics simulation, providing distribution-free prediction intervals with finite-sample coverage guarantees. We further introduce a normalized conformal prediction scheme that leverages MC Dropout uncertainty to produce adaptive-width intervals, yielding tighter intervals in regions of low uncertainty and wider intervals where the model is less certain. Full-scale experiments (33.7M parameters, 800 training samples, 5 ensemble members, NVIDIA V100) on steady-state heat conduction benchmarks demonstrate that our method achieves 89.1% empirical coverage at the target level of alpha=0.1, while producing spatially adaptive prediction intervals that reflect the underlying physical uncertainty structure. We also provide an uncertainty decomposition framework that separates epistemic uncertainty (68% of total) from aleatoric uncertainty (32% of total), offering actionable guidance for data collection and model improvement. Our method is implemented in an open-source platform with REST API endpoints and interactive 3D visualization.

cs.LG↗

arXiv2026

Project Life Cycles in Open-Source Software

Using methods previously applied to product life cycles, this paper models developer engagement through the project life cycle for open-source projects, and detects similar dynamics in a cross section of projects. Endogenous growth theory is used to model growth dynamics in open-source software engineering, while incorporating the interactions between growth levels and developer activity over time using systems of differential equations. The solution to this model calibrates well to many open-source projects. The model generates an estimate of the lifetime developer engagement and growth, which supports estimating a lifetime production value of open-source projects.

cs.SE↗

arXiv2026

Hypothesis-Driven Deep Research with Large Language Models: A Structured Methodology for Automated Knowledge Discovery

Current AI-powered research systems adopt a direct search-then-summarize paradigm that treats hypotheses as end products of scientific discovery. We argue this leaves a critical gap: hypotheses can serve a far more powerful role as organizational instruments that structure the research process itself. We propose the Hypothesis-Driven Deep Research (HDRI) methodology - the first framework using hypotheses to organize general-purpose deep research across arbitrary domains, rather than merely validating claims within specific domains. This transforms research from reactive information retrieval into proactive, verifiable, and iterative knowledge discovery. HDRI is formalized with six core principles and an eight-stage pipeline. A central innovation is the gap-driven iterative research mechanism - a closed-loop quality assurance system that automatically identifies informational and logical gaps, triggering targeted supplementary investigation. We further introduce a fact reasoning framework with traceable reasoning chains and quantified confidence propagation, a subject locking mechanism to prevent entity confusion, and a multi-dimensional quality assessment scheme. The methodology is realized in the INFOMINER system. Experiments demonstrate improvements of 22.4% in fact density, 90% subject matching accuracy, 0.92 multi-source verification confidence, and 14% completeness gain from gap-driven supplementation. Five case studies validate its practical applicability, achieving an average quality rating of 4.46/5.0.

cs.AI↗

arXiv2025

RFSoC Based LLRF System Design at ALS

The Advanced Light Source (ALS) at LBNL is upgrading several LLRF systems for its Linac and Sub-Harmonic Bunchers, where it is desired to have a unified LLRF system design to support various RF frequencies (at 125MHz, 500MHz and 3GHz) and configurations. This paper demonstrates an open-source, direct sampling RFSoC based LLRF system design, featuring: sample-to-sample Multi-Tile Synchronization, deterministic latency, digital up/down conversion, arbitrary waveform generation and acquisition, in-pulse closed loop control, timing and EPICS integration, modular RF frontend and hardware designs. Measured RF characteristics show that the RFSoC based LLRF system is able to meet the system requirements.

physics.acc-ph↗

arXiv2025

Comparative Evaluation of Xilinx RFSoC Platform for Low-Level RF Systems

The rapid advancement of Radio Frequency System-onChip (RFSoC) technology from Xilinx (AMD) has enabled the integration of high-speed data converters and programmable logic within a single package. RFSoC platforms are already widely adopted in telecommunications, radar, and satellite communications, where they promise reductions in system footprint and power consumption. However, their suitability for Low-Level RF (LLRF) control systems in accelerator environments - where stability requirements are critical - has not been quantitatively evaluated. This paper presents a comparative measurement-based assessment of RFSoC-based and conventional LLRF designs, focusing on signal fidelity, phase noise, latency, system complexity, and integration challenges. The advantages and challenges of adopting RFSoC-based direct conversion architectures are discussed, providing guidance for future LLRF system implementations.

physics.acc-ph↗