arXiv subjects

Paul D. Franzon

Publications and source records attributed to Paul D. Franzon.

4 recordsLinked to original sources

arXiv2026

Finding the Needle in a Haystack: Test-Time Analog Circuit Representation Adaptation for Bayesian Optimization

Bayesian optimization (BO) is a sample-efficient framework for analog circuit topology search, where evaluating each candidate topology can require costly simulation. However, representation-based BO methods typically treat circuit embeddings as fixed after encoder training. This creates a mismatch between representation learning and optimization: embeddings learned to encode or reconstruct circuit structure are not necessarily organized according to the figure of merit (FoM) being optimized. This paper introduces Test-Time Analog Representation Adaptation for Bayesian Optimization (TTARO), an online deep-kernel BO framework that adapts circuit representations throughout the search process. Starting from pretrained circuit embeddings, TTARO jointly learns a nonlinear feature transformation and a Gaussian-process surrogate using the FoM labels of the circuits evaluated so far. Following each new evaluation, TTARO updates the representation and surrogate before selecting the next candidate. We compare TTARO with conventional Gaussian Process-based BO over fixed embeddings and with Deep Kernel Learning (DKL), which learns the representation only from the initial evaluated designs and keeps it fixed throughout the remainder of the search. By continually incorporating newly observed FoM labels into representation learning, TTARO aligns the search space with the optimization objective as BO progresses. In our experiments, TTARO reduces regret AUC by 15.2% on average relative to BO and by 20.7% relative to DKL across 40 encoder/kernel/acquisition settings, outperforming prior art in most settings with reductions as large as 46.7%.

cs.LG↗

arXiv2026

Addressing Large Action Spaces in 3D Floorplanning via Spatial Generalization

Many recent machine learning approaches to floorplanning represent placement decisions using discrete canvas coordinates, which creates scalability bottlenecks as the action space grows. In this work, we study the effect of learning a continuous action representation for 3D floorplanning. By reasoning in a continuous placement space and discretizing only at inference time, our method decouples the output structure from the canvas resolution, which makes learning and inference more tractable in large design spaces. A central idea in our approach is \textit{$L$-action similarity}: actions that are close in the placement space often produce similar returns. This smoothness induces a useful structural bias that allows the model to generalize information from one decision to nearby decisions. As a case study, we show that this approach can learn to construct floorplans even when pre-trained only on random floorplans. Our results suggest that continuous decision spaces are a promising way to address the large-action-space challenge in floorplanning.

cs.LG↗

arXiv2025

Adaptive Linear Embedding for Nonstationary High-Dimensional Optimization

Bayesian Optimization (BO) in high-dimensional spaces remains fundamentally limited by the curse of dimensionality and the rigidity of global low-dimensional assumptions. While Random EMbedding Bayesian Optimization (REMBO) mitigates this via linear projections into low-dimensional subspaces, it typically assumes a single global embedding and a stationary objective. In this work, we introduce Self-Adaptive embedding REMBO (SA-REMBO), a novel framework that generalizes REMBO to support multiple random Gaussian embeddings, each capturing a different local subspace structure of the high-dimensional objective. An index variable governs the embedding choice and is jointly modeled with the latent optimization variable via a product kernel in a Gaussian Process surrogate. This enables the optimizer to adaptively select embeddings conditioned on location, effectively capturing locally varying effective dimensionality, nonstationarity, and heteroscedasticity in the objective landscape. We theoretically analyze the expressiveness and stability of the index-conditioned product kernel and empirically demonstrate the advantage of our method across synthetic and real-world high-dimensional benchmarks, where traditional REMBO and other low-rank BO methods fail. Our results establish SA-REMBO as a powerful and flexible extension for scalable BO in complex, structured design spaces.

stat.ML↗

arXiv2024

Can Low-Rank Knowledge Distillation in LLMs be Useful for Microelectronic Reasoning?

In this work, we present empirical results regarding the feasibility of using offline large language models (LLMs) in the context of electronic design automation (EDA). The goal is to investigate and evaluate a contemporary language model's (Llama-2-7B) ability to function as a microelectronic Q & A expert as well as its reasoning, and generation capabilities in solving microelectronic-related problems. Llama-2-7B was tested across a variety of adaptation methods, including introducing a novel low-rank knowledge distillation (LoRA-KD) scheme. Our experiments produce both qualitative and quantitative results.

cs.LG↗