The paper shows that the main limitation of machine‑learning agents is the costly execution step, and proposes a pre‑execution prediction model that achieves 61.5% accuracy, cutting search time by sixfold while improving final performance by about 6%.
A recent ACL 2026 SAC Highlight paper shows that the main limitation of machine‑learning agents is the costly execution step, and demonstrates that large language models can predict which experiment will succeed with 61.5% accuracy, yielding a six‑fold speed‑up in search while improving final performance by 6%.
ml-evolve addresses the shortcomings of generic agent‑search frameworks for machine‑learning pipelines by introducing four specialized agents, staged data gating, and cost‑saving mechanisms, and demonstrates its advantages with a two‑tower retrieval case study and concrete performance metrics.
The article examines how recent advances in large language models, AutoML, and evolutionary algorithms are pushing AI toward recursive self‑improvement, outlines current capabilities and limitations, and discusses the technical, economic, and safety challenges that still prevent a fully autonomous intelligence explosion.
AgenticSciML introduces a collaborative multi‑agent framework that autonomously designs SciML models—such as PINNs and neural operators—by iteratively debating, retrieving knowledge, and evolving solutions, achieving up to 11,000× error reduction on benchmark PDE problems.
UCSD’s AIBuildAI agent achieved first place on OpenAI’s benchmark by automatically designing, coding, training, and tuning a complete AI model without human engineers, a breakthrough that suggests a shift from tool‑assisted AI creation to fully autonomous AI‑generated AI, raising both efficiency gains and new interpretability challenges.
The article examines MiniMax M2.7’s claim of self‑iteration, its impressive Kaggle record, and a series of technical tests—including code refactoring, real‑time chart generation, futures backtesting, business analysis, PPT creation, and news tracking—to evaluate the model’s practical AI self‑evolution capabilities.
Andrej Karpathy released the 630‑line open‑source autoresearch project that lets an AI agent automatically modify training code, run hundreds of short experiments on multiple GPUs, evaluate results, and iteratively improve models, raising questions about whether this constitutes genuine scientific research.
MySQL AI adds built‑in LLMs, embeddings, vector storage, AutoML and a graphical console to on‑premise MySQL, enabling developers to create predictive and generative AI applications—including fraud detection, semantic search, RAG and NL2SQL—without external vector databases or GPUs.
TS‑Agent is a modular LLM‑driven framework that formalizes financial time‑series modeling as a three‑stage iterative decision process, leveraging structured knowledge bases, dynamic memory, and a feedback‑driven code‑editing loop to outperform AutoML baselines in accuracy, robustness, and auditability.
MySQL AI, introduced for the database's 30th anniversary, integrates auto‑ML, generative AI, LLM‑driven text‑to‑SQL, and a vector engine directly into MySQL Enterprise, enabling developers to build intelligent applications on familiar SQL tools without moving data.
This comprehensive guide walks you through generative AI fundamentals—including transformers, diffusion models, large language models, and multimodal systems—then explores practical API usage with OpenAI, Hugging Face, and Vertex AI, followed by model fine‑tuning, LoRA, knowledge injection, and advanced topics such as model distillation, prompt chaining, AutoML, tool integration, and retrieval‑augmented generation.
A recent breakthrough by Tsinghua researchers introduces AutoGFM, an adaptive graph neural architecture search framework that dramatically improves the performance and generalization of graph foundation models across diverse tasks and domains, as validated by extensive ICML‑2025 experiments.
This guide explains how AutoGluon leverages bagging, cross‑validation, and stacked ensembling to automatically train and combine dozens of models, provides step‑by‑step installation and usage instructions for tabular, time‑series, and multimodal tasks, and shows practical deployment examples for industry scenarios.
This article reviews the evolution of AutoML, explains its core principles, compares major Python AutoML libraries with code examples, provides a decision‑making framework and benchmark results, and offers practical guidance on selecting the most suitable tool for different machine‑learning projects.
The JD Exploration Institute paper introduces Omniforce, a human‑centered, cloud‑edge collaborative AutoML system that uses model distillation, dynamic data governance, Bayesian‑optimized training, and edge deployment to cut large‑model training costs by 70% and improve inference speed by 30%, powering the JoyBuild platform for broader AI adoption.
This article explains why Python has become the leading language for data analysis and machine learning, outlines the essential libraries and frameworks, provides practical code examples, describes typical application scenarios, suggests a staged learning roadmap, and forecasts future trends such as AutoML and federated learning.
The article walks through five iterative experiments using MySQL HeatWave AutoML on a golf league dataset, showing how data quality, feature selection, and over‑fitting affect predictive accuracy and emphasizing that high‑quality data is essential for reliable AI models.
This article outlines Tencent's AutoML research, presenting several recent papers that introduce novel neural architecture search, feature selection, pooling, embedding size, and hyper‑parameter optimization techniques to improve the efficiency, accuracy, and scalability of large‑scale advertising recommendation systems.
This article compiles a comprehensive, subjectively curated collection of Python libraries for data visualization, machine learning, deep learning, AutoML, model interpretability, resource monitoring, and debugging, providing brief descriptions and links to each tool for developers and researchers.
This article describes how Du Xiaoman tackled the high cost, instability, and long cycles of AI algorithm deployment by building the ATLAS automated machine learning platform, detailing its four‑stage workflow, component platforms, scaling and efficiency techniques, and practical Q&A for practitioners.
This article surveys recent progress in automated graph machine learning, covering graph neural architecture search techniques such as GASSO, DHGAS, GAUSS, GRACES, and G‑RNA, discusses scalability and robustness challenges, and introduces the open‑source AutoGL library and the NAS‑Bench‑Graph benchmark.
An end‑to‑end AI platform integrates feature processing, model training, deployment, and decision orchestration across offline and online layers, leveraging automated pipelines such as AutoML (feature engineering, hyper‑parameter optimization, neural architecture search) built on Ray Tune and NNI, which have already boosted CTR in real‑world advertising and aim to make every user an algorithm engineer.
This article explains how the Ray distributed computing engine simplifies the design, deployment, and operation of complex cloud‑native distributed systems—illustrated through an AutoML service example—by detailing system complexity, Ray’s core concepts, resource customization, runtime environments, monitoring, and ecosystem integrations.
At the 2023 SECon Global Software Engineering Innovation Summit in Shanghai, Hello’s technology team will showcase how its AI platform leverages AutoML to streamline model development across intelligent mobility services, and how the Quarkc engine uses Web Components to create cross‑stack, framework‑agnostic front‑end components.
This article introduces AutoML, explaining how it automates data cleaning, feature engineering, model selection, hyper‑parameter optimization, and neural architecture search, and showcases Alibaba PAI's implementations of HPO, AutoFE, and NAS with practical case studies and performance results.
This article details Du Xiaoman's development of the ATLAS automated machine learning platform, covering business scenarios, AI algorithm deployment challenges, the end‑to‑end production workflow, platform components such as annotation, data, training and deployment, as well as optimization techniques like AutoML, meta‑learning, NAS, and large‑scale parallelism, concluding with lessons learned and future directions.
This article introduces the EasyModeling framework built on Alibaba Cloud's PAI platform, detailing its modular design, high reusability, integration with deep‑learning libraries, automated hyper‑parameter tuning, deployment scenarios, and a real‑world case study using RoBERTa for dish‑name standardization, demonstrating significant performance gains.
This article introduces OPPO’s self‑developed AutoML platform Pluto, explains why automated machine learning and model compression are essential for industrial AI, describes Pluto’s hardware‑aware and uniform algorithm framework, showcases typical applications such as video super‑resolution, and provides a detailed Q&A on its methodology and performance.
This talk introduces automated knowledge graph representation learning, covering background, key techniques such as triple‑based, path‑based and subgraph‑based models, AutoML‑driven model search (AutoSF, Interstellar, RED‑GNN), evaluation metrics, and future research directions in AI.
This talk introduces the background, key directions, and model designs for automated knowledge‑graph representation learning, covering triple‑based, path‑based, and subgraph‑based approaches, the role of AutoML in searching optimal bilinear scoring functions, and future research challenges such as scalability, inductive inference, and domain‑specific applications.
This article introduces Microsoft’s FLAML, a fast and lightweight AutoML library, explains its design principles, cost‑aware search strategy, key observations, properties, and experimental results, and provides practical code examples for integrating FLAML into Python machine‑learning workflows.
This article presents a comprehensive overview of an intelligent risk control middle‑platform, covering its design background, the five‑characteristics and "five‑all double‑core" concept, full‑cycle strategy and model lifecycle management, business architecture, and real‑world application cases, highlighting the integration of rule‑based and AI‑driven decision engines.
Time series forecasting, essential across weather, finance, and commerce, involves tasks like classification, clustering, anomaly detection, and especially prediction; this article explores its definitions, evaluation metrics, traditional methods, machine‑learning approaches, deep‑learning models such as TFT, and emerging AutoML tools, offering practical insights and best practices.
This article provides an in‑depth overview of the TensorFlow ecosystem, covering Keras modeling productivity tools, classic model examples, AutoKeras and KerasTuner for automated search, data preprocessing pipelines, performance profiling, model optimization, and multiple deployment strategies for servers, browsers, and edge devices.
This article reviews the seven most notable Python libraries and packages of 2018 for data scientists and AI practitioners, including AdaNet, TPOT, SHAP, Optimus, spaCy, Jupytext, and Chartify, with descriptions, installation commands, and usage examples.
We applied knowledge distillation to a video query‑document BERT matcher, compressing the 12‑layer teacher into production‑ready 1‑layer ALBERT and tiny TextCNN students using combined soft, hard, and relevance losses plus AutoML‑tuned hyper‑parameters, achieving sub‑5 ms latency and up to 2.4% AUC improvement over the original model.
This article presents a comprehensive, practical methodology for feature engineering that combines brute‑force AutoML‑style dictionary searches, business‑logic‑driven feature creation, and feature‑importance‑guided refinement, illustrating each approach with real Kaggle competition examples and concrete code snippets.
The article explains how Tuun, an open‑source Bayesian‑optimization tool, automatically searches data‑augmentation policies for machine‑learning models, details the setup with Microsoft NNI, provides code and configuration examples, and presents experiments on CIFAR‑10/100 and SVHN showing that Tuun‑generated policies match or surpass expert‑tuned strategies and further improve performance when combined.
Drawing on seven Kaggle and KDD Cup victories, the article outlines a multi‑domain modeling optimization strategy—covering recommendation, time‑series, and AutoML problems—alongside a three‑module AutoML pipeline and a three‑stage workflow that emphasize systematic evaluation, bias‑variance balance, and robust model‑fusion for competition and industry success.
AutoX offers a one‑click solution for tabular AutoML by defining feature operators, constructing a searchable feature space, applying efficient feature selection, tuning hyper‑parameters, and performing model ensembling, enabling users with limited ML expertise to automatically generate high‑performing predictive models, as demonstrated on multiple Kaggle datasets.
This article presents two ASRU‑2021 accepted works from Kuaishou: MTASS, a multi‑task audio source separation framework that jointly separates speech, music and noise, and SpeechNAS, an AutoML‑based neural architecture search method that achieves state‑of‑the‑art speaker recognition performance with significantly fewer parameters.
Huawei’s 201‑million‑yuan “genius teen” Zhong Zhao leveraged AutoML to deploy high‑precision image‑pixel processing algorithms across tens of millions of Mate and P series smartphones, pioneering large‑scale commercial use of AutoML and advancing mobile visual models with dynamic convolution kernels and adversarial data augmentation.
Huawei unveiled the work of young researcher Zhong Zhao, who within a year applied AutoML to pixel‑level image processing on millions of Mate and P series phones, detailing the technical challenges, novel pipeline, performance gains, and his broader contributions to mobile AI research.
The article presents Akulaku's intelligent pre‑loan risk control framework, detailing a multi‑loop feedback control model, AutoML‑driven controller selection, unsupervised feature extraction techniques, and a comprehensive metric quantification system to improve stability, steady‑state, and dynamic responses of financial risk management.
AutoHERI, a hierarchical representation automatic aggregation model discovered via one‑shot neural architecture search, jointly learns CTR and CVR (and other downstream tasks) to capture cascade relationships, achieving superior AUC and conversion‑rate lifts in large‑scale Alibaba advertising datasets and prompting full production deployment.
This article describes the design, automated workflow, functional modules, and performance results of a comprehensive AI model optimization platform built for Ctrip's travel business, covering operator libraries, graph optimization, model compression techniques such as distillation, quantization, pruning, and deployment integration.
The paper introduces DARTS‑, a robust differentiable architecture search method that adds a linearly decaying auxiliary skip‑connection weight to prevent performance collapse, delivering smoother loss landscapes, lower Hessian spikes, and state‑of‑the‑art accuracy on CIFAR‑10, ImageNet and NAS‑Bench‑201, while maintaining efficiency for large‑scale AutoML deployments.
Ant Group’s financial intelligent AutoML system received the 2020 CCF Technology Advancement Excellence Award, highlighting its industrial‑grade automated modeling algorithms, high‑performance architecture, and large‑scale deployment that boosted AI modeling efficiency by 50% and risk discrimination by 20% in the finance sector.
This article summarizes the Q&A session of the 2020 Tencent Advertising Algorithm Competition live series, covering the fundamentals of automated machine learning, its key technologies, current challenges, and the features and advantages of the SolnML system, while also addressing practical concerns such as hardware support and future research directions.
The team built an automated graph learning framework that preprocesses diverse graphs, employs four GNN architectures, conducts rapid hyper‑parameter tuning, and fuses models with density‑aware weighting, securing first place in KDD Cup 2020 AutoGraph and boosting Meituan’s ad recall and CTR prediction.
This article proposes a meta‑knowledge transfer framework for AutoML systems, detailing a four‑layer architecture, methods for collecting and updating structured model meta‑knowledge, and strategies that use this knowledge to guide hyper‑parameter search and early‑stop training, thereby improving efficiency and reducing resource consumption.
The article presents a comprehensive overview of Autohome's one‑stop machine learning platform, detailing its background, architecture, resource scheduling, data processing, model training (including distributed deep learning), deployment, real‑world applications such as purchase‑intent and recommendation models, and future development directions.
Deepthought is iQIYI’s end‑to‑end machine‑learning platform that unifies distributed frameworks, decouples pipeline stages, integrates with Tongtian Tower, and offers visual drag‑and‑drop configuration, evolving from a fraud‑detection prototype to a generic system with real‑time inference, automated hyper‑parameter optimization, and support for large‑scale data across anti‑fraud, recommendation, and analytics workloads.
This article explains the concept of augmented analytics, compares it with traditional BI, outlines its impact on data preparation, analysis, and machine learning, and reviews the underlying technologies such as NLQ, NLG, AutoML, and data robots, supported by Gartner insights and industry examples.
This article reviews the major challenges of AI—data, model, and knowledge—explains why automated machine learning and neural architecture search are crucial, analyzes weight‑sharing NAS algorithms and their instability, presents various improved DARTS‑based methods, and discusses experimental results and future research directions.
This article surveys the field of Neural Architecture Search (NAS), reviewing its motivations, categorizing methods by search space, search strategy, and performance evaluation, summarizing historical approaches, recent advances, and outlining promising future research directions.
This article presents a comprehensive automated machine‑learning platform that abstracts task design, hyper‑parameter search space management, optimization engines, algorithm repositories, training/evaluation engines, model repositories and monitoring panels, offering both expert‑assisted and code‑free modes to accelerate model building while reducing reliance on specialist knowledge.
In 2020, the AR “scan‑for‑fortune” feature achieved a full AutoML rollout on the xNN‑Cloud platform, automating network architecture design and the entire model development pipeline, which cut Android inference time by over 50%, iOS by 30%, reduced model size, and boosted accuracy by 1.6% while handling billions of in‑client inferences.
This article presents a comprehensive overview of the Automobile Home recommendation system, detailing its objectives, architecture, various ranking models from LR to DeepFM, online learning mechanisms, service APIs, feature engineering pipelines, model training platforms, debugging tools, and future optimization directions.
This article presents a comprehensive study on using automated machine learning (AutoML) to design interaction functions for collaborative filtering, introducing the SIF framework, detailing its search space, one‑shot algorithm, neural architecture search integration, and demonstrating superior performance on benchmark recommendation datasets.
This article provides a comprehensive overview of Automated Machine Learning (AutoML), covering its definition, objectives, research areas, hyperparameter optimization methods, pipeline construction, major CASH algorithms, open-source frameworks such as AutoSklearn and NNI, practical case studies, and current open research challenges.
In this candid interview, Alibaba AI veteran Jia Yangqing shares his late start in programming, the creation of Caffe, perspectives on deep learning, AutoML, knowledge graphs, and practical advice for engineers on career growth, research focus, and the challenges of bringing AI to real‑world products.
This article reviews reinforcement‑learning‑driven neural architecture search, covering layer‑based, block‑based, and connection‑based strategies, as well as advanced techniques such as inverse reinforcement learning, graph hyper‑networks, Monte‑Carlo tree search, and knowledge‑distillation‑based model compression.
The talk outlines how Tencent Advertising leverages deep‑learning visual algorithms—including GCN‑based edge refinement, template generation, AutoML‑driven smart review, and a dual‑tower click‑through‑rate model—to automate creative production, improve ad quality, and enhance user experience across creation, review, and playback stages.
This article reviews Huawei's ModelArts platform, detailing its data processing, algorithm development, high‑performance training, edge‑cloud model deployment, auto‑learning capabilities, and real‑world use cases such as invisible payment and intelligent waste classification, while outlining future ecosystem prospects.
This article provides a comprehensive overview of AutoML, describing its motivation, formal definition, typical machine‑learning pipeline, key challenges, various optimizer and evaluator strategies—including simple search, heuristic, model‑based, reinforcement learning, and meta‑learning approaches—along with practical applications and future prospects.
This talk presents the research and practical deployment of AutoML for tabular data, covering background, automated feature engineering and selection, hyper‑parameter optimization, the AutoCross feature‑crossing system, case studies, and future directions, demonstrating its advantages over Google Cloud AutoML on multiple Kaggle competitions.
This article presents a comprehensive overview of applying Automated Machine Learning (AutoML) to recommendation systems, detailing methods for data preprocessing, feature engineering, model selection, hyper‑parameter optimization, and neural architecture search, and shares practical experiences and performance gains observed in real‑world deployments.
Alibaba’s AI Lab introduces AutoAI, a fully managed AutoML platform that automates data production, model training, and deployment, reducing human effort, improving efficiency, ensuring data security, and enabling one‑click model training across NLP, speech, and vision applications.
This article reviews Jia Yangqing’s insights on why deep learning finally succeeded—highlighting the roles of big data and high‑performance computing—while examining its current limitations, emerging challenges, and future opportunities across AI engineering, AutoML, and hardware‑software co‑design.
This article explains what hyper‑parameters are, why their tuning is a black‑box problem, and how Bayesian optimization—using surrogate models, acquisition functions, and posterior inference—offers a more efficient alternative to grid or random search, while also listing popular open‑source tools and discussing its limitations.
This article introduces the open‑source H2O platform, walks through installing the Python package, demonstrates data import, model building with GBM and AutoML, evaluates results, explains model deployment via POJO/MOJO, and discusses the visual Flow UI and broader implications of automated modeling.
This article introduces the open‑source H2O platform, explains how to install and use its Python API for data loading, preprocessing, model training with GBM and AutoML, evaluates results with AUC, and describes model deployment via POJO/MOJO as well as the visual Flow UI, concluding with reflections on the role of automated modeling in data science.
AutoML and Neural Architecture Search automate deep‑learning model design by using controller networks to explore and evaluate candidate architectures, with efficient variants like PNAS and ENAS reducing cost, while platforms such as Google Cloud AutoML and open‑source AutoKeras make these techniques accessible, promising broader, democratized AI breakthroughs.
AutoML and Neural Architecture Search automate deep‑learning model design by sampling and training network blocks, using reinforcement‑learning or efficient weight‑sharing strategies such as PNAS and ENAS, enabling high‑accuracy architectures in days on a single GPU, with services like Google Cloud AutoML and open‑source tools like AutoKeras, while future research aims to expand search spaces beyond hand‑crafted blocks.
The article explains how large‑scale machine learning and AutoML are applied to search advertising click‑through‑rate (CTR) prediction, covering problem definition, feature generation, model training, optimization methods, distributed systems, and recent advances in AutoML with practical case studies.
At the AI Pioneer Conference, Wang Xiaobo, head of Alibaba’s Commercial Machine Intelligence and Youku’s algorithm teams, discussed large‑scale distributed learning, recommendation challenges such as cold‑start and video heterogeneity, AutoML innovations, multi‑modal search during promotions, and the future demand for specialists in few‑shot learning and domain adaptation.
Google Cloud NEXT 18 in San Francisco unveiled the alpha‑tested Cloud TPU 3.0, major AutoML enhancements, and the Contact Center AI solution, while CEO Diane Greene highlighted AI and security investments and the cloud’s rapid revenue growth, signaling Google’s push to outpace AWS, Azure, and IBM.
In this interview, Long Mingkang, Vice President of iFlytek's Cloud Computing Institute, shares his experience building large‑scale speech cloud services, discusses the technical hurdles of speech recognition and AI platform development, compares TensorFlow and MXNet, and offers insights on AutoML, industry trends, and how engineers can master AI.
Google’s AutoML system has begun generating machine‑learning code that outperforms the very researchers who created it, achieving record‑high accuracy on image‑recognition tasks and demonstrating a self‑replicating AI that challenges the notion of human superiority in programming.
Google's AutoML system can automatically generate and improve machine‑learning code, outperforming human researchers with record‑high accuracy on image‑recognition tasks and demonstrating that AI‑driven self‑replicating programs can surpass programmers in just a few hours.
