Abstract:Recent advances in machine learning (ML) for automating analog circuit synthesis have been significant, yet challenges remain. A critical gap is the lack of a standardized evaluation framework, compounded by various process design kits (PDKs), simulation tools, and a limited variety of circuit topologies. These factors hinder direct comparisons and the validation of algorithms. To address these shortcomings, we introduced AnalogGym, an open-source testing suite designed to provide fair and comprehensive evaluations. AnalogGym includes 30 circuit topologies in five categories: sensing front ends, voltage references, low dropout regulators, amplifiers, and phase-locked loops. It supports several technology nodes for academic and commercial applications and is compatible with commercial simulators such as Cadence Spectre, Synopsys HSPICE, and the open-source simulator Ngspice. AnalogGym standardizes the assessment of ML algorithms in analog circuit synthesis and promotes reproducibility with its open datasets and detailed benchmark specifications. AnalogGym's user-friendly design allows researchers to easily adapt it for robust, transparent comparisons of state-of-the-art methods, while also exposing them to real-world industrial design challenges, enhancing the practical relevance of their work. Additionally, we have conducted a comprehensive comparison study of various analog sizing methods on AnalogGym, highlighting the capabilities and advantages of different approaches. AnalogGym is available in the GitHub repository <a class="link-external link-https" href="https://github.com/CODA-Team/AnalogGym" rel="external noopener nofollow">this https URL</a>. The documentation is also available at <a class="link-external link-http" href="http://coda-team.github.io/AnalogGym/" rel="external noopener nofollow">this http URL</a>.

OpenLS-DGF: An Adaptive Open-Source Dataset Generation Framework for Machine Learning Tasks in Logic Synthesis

GPT-LS: Generative Pre-Trained Transformer with Offline Reinforcement Learning for Logic Synthesis.

HLSDataset: Open-Source Dataset for ML-Assisted FPGA Design using High Level Synthesis

Data is all you need: Finetuning LLMs for Chip Design via an Automated design-data augmentation framework

LOSTIN: Logic Optimization Via Spatio-Temporal Information with Hybrid Graph Models

Logic Synthesis Optimization with Predictive Self-Supervision via Causal Transformers

ERL-LS: Accelerating Primitive Sequence Generation of Logic Synthesis with Evolutionary Reinforcement Learning

HLSFactory: A Framework Empowering High-Level Synthesis Datasets for Machine Learning and Beyond

INVICTUS: Optimizing Boolean Logic Circuit Synthesis via Synergistic Learning and Search

LOSTIN: (L)under-barogic (O)under-barptimization via (S)under-barpatio-(T)under-baremporal (In)under-barformation with Hybrid Graph Models

MG-Verilog: Multi-grained Dataset Towards Enhanced LLM-assisted Verilog Generation

CircuitNet: an Open-Source Dataset for Machine Learning in VLSI CAD Applications with Improved Domain-Specific Evaluation Metric and Learning Strategies.

A Circuit Domain Generalization Framework for Efficient Logic Synthesis in Chip Design

CircuitNet: An Open-Source Dataset for Machine Learning Applications in Electronic Design Automation (EDA)

MTLSO: A Multi-Task Learning Approach for Logic Synthesis Optimization

AnalogGym: An Open and Practical Testing Suite for Analog Circuit Synthesis

LSOracle: a Logic Synthesis Framework Driven by Artificial Intelligence: Invited Paper

DeLoSo: Detecting Logic Synthesis Optimization Faults Based on Configuration Diversity

A General Framework for Efficient Logic Synthesis

Too Big to Fail? Active Few-Shot Learning Guided Logic Synthesis

OpenR: An Open Source Framework for Advanced Reasoning with Large Language Models