Jianfeng Zhan

Yatao Li, Wanling Gao, Lei Wang et al.

AI for science (AI4S) is an emerging research field that aims to enhance the accuracy and speed of scientific computing tasks using machine learning methods. Traditional AI benchmarking methods struggle to adapt to the unique challenges posed by AI4S because they assume data in training, testing, and future real-world queries are independent and identically distributed, while AI4S workloads anticipate out-of-distribution problem instances. This paper investigates the need for a novel approach to effectively benchmark AI for science, using the machine learning force field (MLFF) as a case study. MLFF is a method to accelerate molecular dynamics (MD) simulation with low computational cost and high accuracy. We identify various missed opportunities in scientifically meaningful benchmarking and propose solutions to evaluate MLFF models, specifically in the aspects of sample efficiency, time domain sensitivity, and cross-dataset generalization capabilities. By setting up the problem instantiation similar to the actual scientific applications, more meaningful performance metrics from the benchmark can be achieved. This suite of metrics has demonstrated a better ability to assess a model's performance in real-world scientific applications, in contrast to traditional AI benchmarking methodologies. This work is a component of the SAIBench project, an AI4S benchmarking suite. The project homepage is https://www.computercouncil.org/SAIBench.

Xu Wen, Wanling Gao, Anzheng Li et al.

Classical machine learning (CML) occupies nearly half of machine learning pipelines in production applications. Unfortunately, it fails to utilize the state-of-the-practice devices fully and performs poorly. Without a unified framework, the hybrid deployments of deep learning (DL) and CML also suffer from severe performance and portability issues. This paper presents the design of a unified compiler, called CMLCompiler, for CML inference. We propose two unified abstractions: operator representations and extended computational graphs. The CMLCompiler framework performs the conversion and graph optimization based on two unified abstractions, then outputs an optimized computational graph to DL compilers or frameworks. We implement CMLCompiler on TVM. The evaluation shows CMLCompiler's portability and superior performance. It achieves up to 4.38$\times$ speedup on CPU, 3.31$\times$ speedup on GPU, and 5.09$\times$ speedup on IoT devices, compared to the state-of-the-art solutions -- scikit-learn, intel sklearn, and hummingbird. Our performance of CML and DL mixed pipelines achieves up to 3.04x speedup compared with cross-framework implementations. The project documents and source code are available at https://www.computercouncil.org/cmlcompiler.

Zhengxin Yang, Wanling Gao, Chunjie Luo et al.

Machine learning inference should be subject to stringent inference time constraints while ensuring high inference quality, especially in safety-critical (e.g., autonomous driving) and mission-critical (e.g., emotion recognition) contexts. Neglecting either aspect can lead to severe consequences, such as loss of life and property damage. Many studies lack a comprehensive consideration of these metrics, leading to incomplete or misleading evaluations. The study unveils a counterintuitive revelation: deep learning inference quality exhibits fluctuations due to inference time. To depict this phenomenon, the authors coin a new term, "tail quality," providing a more comprehensive evaluation, and overcoming conventional metric limitations. Moreover, the research proposes an initial evaluation framework to analyze factors affecting quality fluctuations, facilitating the prediction of the potential distribution of inference quality. The effectiveness of the evaluation framework is validated through experiments conducted on deep learning models for three different tasks across four systems.

Hongxiao Li, Wanling Gao, Lei Wang et al.

\textsc{Lara} is a key-value algebra that aims at unifying linear and relational algebra with three types of operation abstraction. The study of \textsc{Lara}'s expressive ability reports that it can represent relational algebra and most linear algebra operations. However, several essential computations, such as matrix inversion and determinant, cannot be expressed in \textsc{Lara}. \textsc{Lara} cannot represent global and iterative computation, either. This article proposes \textsc{IterLara}, extending \textsc{Lara} with iterative operators, to provide an algebraic model that unifies operations in general-purpose computing, like big data, AI, scientific computing, and database. We study the expressive ability of \textsc{Lara} and \textsc{IterLara} and prove that \textsc{IterLara} with aggregation functions can represent matrix inversion, determinant. Besides, we demonstrate that \textsc{IterLara} with no limitation of function utility is Turing complete. We also propose the Operation Count (OP) as a metric of computation amount for \textsc{IterLara} and ensure that the OP metric is in accordance with the existing computation metrics.

Fanda Fan, Chaoxu Guo, Litong Gong et al.

Video outpainting aims to adequately complete missing areas at the edges of video frames. Compared to image outpainting, it presents an additional challenge as the model should maintain the temporal consistency of the filled area. In this paper, we introduce a masked 3D diffusion model for video outpainting. We use the technique of mask modeling to train the 3D diffusion model. This allows us to use multiple guide frames to connect the results of multiple video clip inferences, thus ensuring temporal consistency and reducing jitter between adjacent frames. Meanwhile, we extract the global frames of the video as prompts and guide the model to obtain information other than the current video clip using cross-attention. We also introduce a hybrid coarse-to-fine inference pipeline to alleviate the artifact accumulation problem. The existing coarse-to-fine pipeline only uses the infilling strategy, which brings degradation because the time interval of the sparse frames is too large. Our pipeline benefits from bidirectional learning of the mask modeling and thus can employ a hybrid strategy of infilling and interpolation when generating sparse frames. Experiments show that our method achieves state-of-the-art results in video outpainting tasks. More results and codes are provided at our https://fanfanda.github.io/M3DDM/.

Yatao Li, Jianfeng Zhan

Scientific research communities are embracing AI-based solutions to target tractable scientific tasks and improve research workflows. However, the development and evaluation of such solutions are scattered across multiple disciplines. We formalize the problem of scientific AI benchmarking, and propose a system called SAIBench in the hope of unifying the efforts and enabling low-friction on-boarding of new disciplines. The system approaches this goal with SAIL, a domain-specific language to decouple research problems, AI models, ranking criteria, and software/hardware configuration into reusable modules. We show that this approach is flexible and can adapt to problems, AI models, and evaluation methods defined in different perspectives. The project homepage is https://www.computercouncil.org/SAIBench

Chenxi Wang, Lei Wang, Wanling Gao et al.

In a computer system, multiple indispensable components-such as the CPU, memory, and others-work together with other essential components to produce an overall effect, which can only be measured on an independently running system. Since the system operates as an integrated whole, isolating the effect of individual components is challenging. Accurately attributing the system's overall effect to its specific component is crucial for both computer design and evaluation. Taking CPU evaluation as a benchmark, our experiments reveal that the general-purpose rigorous methodologies, like DoE, RCTs, can not address this issue efficiently; A single-purpose empirical methodology, SPEC CPU2017, which is the industry-standard CPU benchmark, only reports the overall effect. Even more concerningly, for the identical CPU, the undefined configurations of other indispensable components introduce uncontrolled variability, with the SPEC scores fluctuating from 12.16\% to 436.80\%. We propose a rigorous methodology that can attribute the overall effect to its specific component, which can be utilized in computer component evaluations and design, as well as in other areas. Through theoretical analysis and pioneering controlled experiments, we systematically compare our methodology against three established methodologies: SPEC CPU2017, DoE, and RCTs. The results show our methodology can achieve its goal in a cost-efficient way, while others exhibit inherent limitations.

Fei Tang, Wanling Gao, Luzhou Peng et al.

Large language models (LLMs) like ChatGPT have revealed amazing intelligence. How to evaluate the question-solving abilities of LLMs and their degrees of intelligence is a hot-spot but challenging issue. First, the question-solving abilities are interlaced with different ability branches like understanding and massive knowledge categories like mathematics. Second, the inputs of questions are multimodal that may involve text and images. Third, the response format of LLMs is diverse and thus poses great challenges for result extraction and evaluation. In this paper, we propose AGIBench -- a multi-granularity, multimodal, human-referenced, and auto-scoring benchmarking methodology for LLMs. Instead of a collection of blended questions, AGIBench focuses on three typical ability branches and adopts a four-tuple <ability branch, knowledge, difficulty, modal> to label the attributes of each question. First, it supports multi-granularity benchmarking, e.g., per-question, per-ability branch, per-knowledge, per-modal, per-dataset, and per-difficulty level granularities. Second, it contains multimodal input, including text and images. Third, it classifies all the questions into five degrees of difficulty according to the average accuracy rate of abundant educated humans (human-referenced). Fourth, it adopts zero-shot learning to avoid introducing additional unpredictability and provides an auto-scoring method to extract and judge the result. Finally, it defines multi-dimensional metrics, including accuracy under the average, worst, best, and majority voting cases, and repeatability. AGIBench is publically available from \url{https://www.benchcouncil.org/agibench}.

Hongxiao Li, Wanling Gao, Lei Wang et al.

Previous computation models either have equivalent abilities in representing all computations but fail to provide primitive operators for programming complex algorithms or lack generalized expression ability to represent newly-added computations. This article presents a unified computation model with generalized expression ability and a concise set of primitive operators for programming high-level algorithms. We propose a unified data abstraction -- Tensor of List, and offer a unified computation model based on Tensor of List, which we call the ToL model (in short, ToL). ToL introduces five atomic computations that can represent any elementary computation by finite composition, ensured with strict formal proof. Based on ToL, we design a pure-functional language -- ToLang. ToLang provides a concise set of primitive operators that can be used to program complex big data and AI algorithms. Our evaluations show ToL has generalized expression ability and a built-in performance indicator, born with a strictly defined computation metric -- elementary operation count (EOPs), consistent with FLOPs within a small error range.

Xiaorui Wang, Fanda Fan, Chenxi Wang et al.

Recent progress in time-series forecasting has led to rapidly increasing architectural complexity, yet many reported State-of-the-Art gains are statistically fragile or misattributed. We argue that progress requires a shift from model selection to modular attribution, identifying which components truly drive performance. We propose CombinationTS, a self-contained probabilistic evaluation framework that decomposes forecasting models into orthogonal modules--Input Transformation, Embedding, Encoder, Decoder, and Output Transformation--and evaluates them under a shared evaluation condition space. By quantifying each component via marginalized performance ($μ$) and stability ($σ$), CombinationTS enables robust attribution beyond fragile point estimates. Through large-scale paired evaluation, we uncover the Identity Paradox: once the data view (Embedding) is well-designed, a parameter-free Identity Encoder often matches or outperforms complex backbones. We further show that explicit structural priors introduced via Input Transformations yield a more favorable performance-stability trade-off than increasing Encoder complexity, establishing a principled baseline for architectural necessity.

Fanda Fan, Chunjie Luo, Wanling Gao et al.

The burgeoning field of Artificial Intelligence Generated Content (AIGC) is witnessing rapid advancements, particularly in video generation. This paper introduces AIGCBench, a pioneering comprehensive and scalable benchmark designed to evaluate a variety of video generation tasks, with a primary focus on Image-to-Video (I2V) generation. AIGCBench tackles the limitations of existing benchmarks, which suffer from a lack of diverse datasets, by including a varied and open-domain image-text dataset that evaluates different state-of-the-art algorithms under equivalent conditions. We employ a novel text combiner and GPT-4 to create rich text prompts, which are then used to generate images via advanced Text-to-Image models. To establish a unified evaluation framework for video generation tasks, our benchmark includes 11 metrics spanning four dimensions to assess algorithm performance. These dimensions are control-video alignment, motion effects, temporal consistency, and video quality. These metrics are both reference video-dependent and video-free, ensuring a comprehensive evaluation strategy. The evaluation standard proposed correlates well with human judgment, providing insights into the strengths and weaknesses of current I2V algorithms. The findings from our extensive experiments aim to stimulate further research and development in the I2V field. AIGCBench represents a significant step toward creating standardized benchmarks for the broader AIGC landscape, proposing an adaptable and equitable framework for future assessments of video generation tasks. We have open-sourced the dataset and evaluation code on the project website: https://www.benchcouncil.org/AIGCBench.

Wanling Gao, Yunyou Huang, Dandan Cui et al.

A profound gap persists between artificial intelligence (AI) and clinical practice in medicine, primarily due to the lack of rigorous and cost-effective evaluation methodologies. State-of-the-art and state-of-the-practice AI model evaluations are limited to laboratory studies on medical datasets or direct clinical trials with no or solely patient-centered controls. Moreover, the crucial role of clinicians in collaborating with AI, pivotal for determining its impact on clinical practice, is often overlooked. For the first time, we emphasize the critical necessity for rigorous and cost-effective evaluation methodologies for AI models in clinical practice, featuring patient/clinician-centered (dual-centered) AI randomized controlled trials (DC-AI RCTs) and virtual clinician-based in-silico trials (VC-MedAI) as an effective proxy for DC-AI RCTs. Leveraging 7500 diagnosis records from two-step inaugural DC-AI RCTs across 14 medical centers with 125 clinicians, our results demonstrate the necessity of DC-AI RCTs and the effectiveness of VC-MedAI. Notably, VC-MedAI performs comparably to human clinicians, replicating insights and conclusions from prospective DC-AI RCTs. We envision DC-AI RCTs and VC-MedAI as pivotal advancements, presenting innovative and transformative evaluation methodologies for AI models in clinical practice, offering a preclinical-like setting mirroring conventional medicine, and reshaping development paradigms in a cost-effective and fast-iterative manner. Chinese Clinical Trial Registration: ChiCTR2400086816.

Yikang Yang, Zhengxin Yang, Minghao Luo et al.

Finding frequently occurring subgraph patterns or network motifs in neural architectures is crucial for optimizing efficiency, accelerating design, and uncovering structural insights. However, as the subgraph size increases, enumeration-based methods are perfectly accurate but computationally prohibitive, while sampling-based methods are computationally tractable but suffer from a severe decline in discovery capability. To address these challenges, this paper proposes GraDE, a diffusion-guided search framework that ensures both computational feasibility and discovery capability. The key innovation is the Graph Diffusion Estimator (GraDE), which is the first to introduce graph diffusion models to identify frequent subgraphs by scoring their typicality within the learned distribution. Comprehensive experiments demonstrate that the estimator achieves superior ranking accuracy, with up to 114\% improvement compared to sampling-based baselines. Benefiting from this, the proposed framework successfully discovers large-scale frequent patterns, achieving up to 30$\times$ higher median frequency than sampling-based methods.

Yunyou Huang, Xiaoshuang Liang, Xiangjiang Lu et al.

Although Alzheimer's disease (AD) cannot be reversed or cured, timely diagnosis can significantly reduce the burden of treatment and care. Current research on AD diagnosis models usually regards the diagnosis task as a typical classification task with two primary assumptions: 1) All target categories are known a priori; 2) The diagnostic strategy for each patient is consistent, that is, the number and type of model input data for each patient are the same. However, real-world clinical settings are open, with complexity and uncertainty in terms of both subjects and the resources of the medical institutions. This means that diagnostic models may encounter unseen disease categories and need to dynamically develop diagnostic strategies based on the subject's specific circumstances and available medical resources. Thus, the AD diagnosis task is tangled and coupled with the diagnosis strategy formulation. To promote the application of diagnostic systems in real-world clinical settings, we propose OpenClinicalAI for direct AD diagnosis in complex and uncertain clinical settings. This is the first powerful end-to-end model to dynamically formulate diagnostic strategies and provide diagnostic results based on the subject's conditions and available medical resources. OpenClinicalAI combines reciprocally coupled deep multiaction reinforcement learning (DMARL) for diagnostic strategy formulation and multicenter meta-learning (MCML) for open-set recognition. The experimental results show that OpenClinicalAI achieves better performance and fewer clinical examinations than the state-of-the-art model. Our method provides an opportunity to embed the AD diagnostic system into the current health care system to cooperate with clinicians to improve current health care.

Yunyou Huang, Xianglong Guan, Xiangjiang Lu et al.

Alzheimer's disease (AD) cannot be reversed, but early diagnosis will significantly benefit patients' medical treatment and care. In recent works, AD diagnosis has the primary assumption that all categories are known a prior -- a closed-set classification problem, which contrasts with the open-set recognition problem. This assumption hinders the application of the model in natural clinical settings. Although many open-set recognition technologies have been proposed in other fields, they are challenging to use for AD diagnosis directly since 1) AD is a degenerative disease of the nervous system with similar symptoms at each stage, and it is difficult to distinguish from its pre-state, and 2) diversified strategies for AD diagnosis are challenging to model uniformly. In this work, inspired by the concerns of clinicians during diagnosis, we propose an open-set recognition model, OpenAPMax, based on the anomaly pattern to address AD diagnosis in real-world settings. OpenAPMax first obtains the abnormal pattern of each patient relative to each known category through statistics or a literature search, clusters the patients' abnormal pattern, and finally, uses extreme value theory (EVT) to model the distance between each patient's abnormal pattern and the center of their category and modify the classification probability. We evaluate the performance of the proposed method with recent open-set recognition, where we obtain state-of-the-art results.

Yatao Li, Jianfeng Zhan

Artificial Intelligence for Science (AI4S) is an emerging research field that utilizes machine learning advancements to tackle complex scientific computational issues, aiming to enhance computational efficiency and accuracy. However, the data-driven nature of AI4S lacks the correctness or accuracy assurances of conventional scientific computing, posing challenges when deploying AI4S models in real-world applications. To mitigate these, more comprehensive benchmarking procedures are needed to better understand AI4S models. This paper introduces a novel benchmarking approach, known as structural interpretation, which addresses two key requirements: identifying the trusted operating range in the problem space and tracing errors back to their computational components. This method partitions both the problem and metric spaces, facilitating a structural exploration of these spaces. The practical utility and effectiveness of structural interpretation are illustrated through its application to three distinct AI4S workloads: machine-learning force fields (MLFF), jet tagging, and precipitation nowcasting. The benchmarks effectively model the trusted operating range, trace errors, and reveal novel perspectives for refining the model, training process, and data sampling strategy. This work is part of the SAIBench project, an AI4S benchmarking suite.

Yuan Liang, Yange Guo, Yanxia Gong et al.

Federated learning is a new machine learning paradigm. The goal is to build a machine learning model from the data sets distributed on multiple devices so-called an isolated data island, while keeping their data secure and private. Most existing federated learning benchmarks work manually splits commonly used public datasets into partitions to simulate real world isolated data island scenarios. Still, this simulation fails to capture real world isolated data island intrinsic characteristics. This paper presents a federated learning (FL) benchmark suite named FLBench. FLBench contains three domains: medical, financial, and AIoT. By configuring various domains, FLBench is qualified to evaluate federated learning systems and algorithms essential aspects, like communication, scenario transformation, privacy-preserving, data distribution heterogeneity, and cooperation strategy. Hence, it becomes a promising platform for developing novel federated learning algorithms. Currently, FLBench is open sourced and in fast evolution. We package it as an automated deployment tool. The benchmark suite is available from https://www.benchcouncil.org/flbench.html.

Zihan Jiang, Wanling Gao, Lei Wang et al.

In recent years, with the trend of applying deep learning (DL) in high performance scientific computing, the unique characteristics of emerging DL workloads in HPC raise great challenges in designing, implementing HPC AI systems. The community needs a new yard stick for evaluating the future HPC systems. In this paper, we propose HPC AI500 --- a benchmark suite for evaluating HPC systems that running scientific DL workloads. Covering the most representative scientific fields, each workload from HPC AI500 is based on real-world scientific DL applications. Currently, we choose 14 scientific DL benchmarks from perspectives of application scenarios, data sets, and software stack. We propose a set of metrics for comprehensively evaluating the HPC AI systems, considering both accuracy, performance as well as power and cost. We provide a scalable reference implementation of HPC AI500. HPC AI500 is a part of the open-source AIBench project, the specification and source code are publicly available from \url{http://www.benchcouncil.org/AIBench/index.html}.

Wanling Gao, Yuqing Zhu, Zhen Jia et al.

This paper presents our joint research efforts on big data benchmarking with several industrial partners. Considering the complexity, diversity, workload churns, and rapid evolution of big data systems, we take an incremental approach in big data benchmarking. For the first step, we pay attention to search engines, which are the most important domain in Internet services in terms of the number of page views and daily visitors. However, search engine service providers treat data, applications, and web access logs as business confidentiality, which prevents us from building benchmarks. To overcome those difficulties, with several industry partners, we widely investigated the open source solutions in search engines, and obtained the permission of using anonymous Web access logs. Moreover, with two years' great efforts, we created a sematic search engine named ProfSearch (available from http://prof.ict.ac.cn). These efforts pave the path for our big data benchmark suite from search engines---BigDataBench, which is released on the web page (http://prof.ict.ac.cn/BigDataBench). We report our detailed analysis of search engine workloads, and present our benchmarking methodology. An innovative data generation methodology and tool are proposed to generate scalable volumes of big data from a small seed of real data, preserving semantics and locality of data. Also, we preliminarily report two case studies using BigDataBench for both system and architecture researches.

Kuiye Ding, Fanda Fan, Yao Wang et al.

Multivariate Time Series Forecasting plays a key role in many applications. Recent works have explored using Large Language Models for MTSF to take advantage of their reasoning abilities. However, many methods treat LLMs as end-to-end forecasters, which often leads to a loss of numerical precision and forces LLMs to handle patterns beyond their intended design. Alternatively, methods that attempt to align textual and time series modalities within latent space frequently encounter alignment difficulty. In this paper, we propose to treat LLMs not as standalone forecasters, but as semantic guidance modules within a dual-stream framework. We propose DualSG, a dual-stream framework that provides explicit semantic guidance, where LLMs act as Semantic Guides to refine rather than replace traditional predictions. As part of DualSG, we introduce Time Series Caption, an explicit prompt format that summarizes trend patterns in natural language and provides interpretable context for LLMs, rather than relying on implicit alignment between text and time series in the latent space. We also design a caption-guided fusion module that explicitly models inter-variable relationships while reducing noise and computation. Experiments on real-world datasets from diverse domains show that DualSG consistently outperforms 15 state-of-the-art baselines, demonstrating the value of explicitly combining numerical forecasting with semantic guidance.

Kuiye Ding, Fanda Fan, Chunyi Hou et al.

Multivariate time series forecasting is essential in domains such as finance, transportation, climate, and energy. However, existing patch-based methods typically adopt fixed-length segmentation, overlooking the heterogeneity of local temporal dynamics and the decoding heterogeneity of forecasting. Such designs lose details in information-dense regions, introduce redundancy in stable segments, and fail to capture the distinct complexities of short-term and long-term horizons. We propose TimeMosaic, a forecasting framework that aims to address temporal heterogeneity. TimeMosaic employs adaptive patch embedding to dynamically adjust granularity according to local information density, balancing motif reuse with structural clarity while preserving temporal continuity. In addition, it introduces segment-wise decoding that treats each prediction horizon as a related subtask and adapts to horizon-specific difficulty and information requirements, rather than applying a single uniform decoder. Extensive evaluations on benchmark datasets demonstrate that TimeMosaic delivers consistent improvements over existing methods, and our model trained on the large-scale corpus with 321 billion observations achieves performance competitive with state-of-the-art TSFMs.

Hongxiao Li, Chenxi Wang, Fanda Fan et al.

Evaluation is the foundation of empirical science, yet the evaluation of evaluation itself -- so-called meta-evaluation -- remains strikingly underdeveloped. While methods such as observational studies, design of experiments (DoE), and randomized controlled trials (RCTs) have shaped modern scientific practice, there has been little systematic inquiry into their comparative validity and utility across domains. Here we introduce a formal framework for meta-evaluation by defining the evaluation space, its structured representation, and a benchmark we call AxiaBench. AxiaBench enables the first large-scale, quantitative comparison of ten widely used evaluation methods across eight representative application domains. Our analysis reveals a fundamental limitation: no existing method simultaneously achieves accuracy and efficiency across diverse scenarios, with DoE and observational designs in particular showing significant deviations from real-world ground truth. We further evaluate a unified method of entire-space stratified sampling from previous evaluatology research, and the results report that it consistently outperforms prior approaches across all tested domains. These results establish meta-evaluation as a scientific object in its own right and provide both a conceptual foundation and a pragmatic tool set for advancing trustworthy evaluation in computational and experimental research.

Kuiye Ding, Fanda Fan, Zheya Wang et al.

In the World Wide Web, reliable time series forecasts provide the forward-looking signals that drive resource planning, cache placement, and anomaly response, enabling platforms to operate efficiently as user behavior and content distributions evolve. Compared with other domains, time series forecasting for Web applications requires much faster responsiveness to support real-time decision making. We present KAIROS, a non-autoregressive time series forecasting framework that directly models segment-level multi-peak distributions. Unlike autoregressive approaches, KAIROS avoids error accumulation and achieves just-in-time inference, while improving over existing non-autoregressive models that collapse to over-smoothed predictions. Trained on the large-scale corpus, KAIROS demonstrates strong zero-shot generalization on six widely used benchmarks, delivering forecasting performance comparable to state-of-the-art foundation models with similar scale, at a fraction of their inference cost. Beyond empirical results, KAIROS highlights the importance of non-autoregressive design as a scalable paradigm for foundation models in time series.

Zhengxin Yang, Wanling Gao, Luzhou Peng et al.

Designing and optimizing neural network architectures typically requires extensive expertise, starting with handcrafted designs and then manual or automated refinement. This dependency presents a significant barrier to rapid innovation. Recognizing the complexity of automatically generating neural network architecture from scratch, we introduce Younger, a pioneering dataset to advance this ambitious goal. Derived from over 174K real-world models across more than 30 tasks from various public model hubs, Younger includes 7,629 unique architectures, and each is represented as a directed acyclic graph with detailed operator-level information. The dataset facilitates two primary design paradigms: global, for creating complete architectures from scratch, and local, for detailed architecture component refinement. By establishing these capabilities, Younger contributes to a new frontier, Artificial Intelligence-Generated Neural Network Architecture (AIGNNA). Our experiments explore the potential and effectiveness of Younger for automated architecture generation and, as a secondary benefit, demonstrate that Younger can serve as a benchmark dataset, advancing the development of graph neural networks. We release the dataset and code publicly to lower the entry barriers and encourage further research in this challenging area.

Yunyou Huang, Nana Wang, Suqin Tang et al.

This paper quantitatively reveals the state-of-the-art and state-of-the-practice AI systems only achieve acceptable performance on the stringent conditions that all categories of subjects are known, which we call closed clinical settings, but fail to work in real-world clinical settings. Compared to the diagnosis task in the closed setting, real-world clinical settings pose severe challenges, and we must treat them differently. We build a clinical AI benchmark named Clinical AIBench to set up real-world clinical settings to facilitate researches. We propose an open, dynamic machine learning framework and develop an AI system named OpenClinicalAI to diagnose diseases in real-world clinical settings. The first versions of Clinical AIBench and OpenClinicalAI target Alzheimer's disease. In the real-world clinical setting, OpenClinicalAI significantly outperforms the state-of-the-art AI system. In addition, OpenClinicalAI develops personalized diagnosis strategies to avoid unnecessary testing and seamlessly collaborates with clinicians. It is promising to be embedded in the current medical systems to improve medical services.

Chunjie Luo, Jianfeng Zhan, Tianshu Hao et al.

Attention is an effective mechanism to improve the deep model capability. Squeeze-and-Excite (SE) introduces a light-weight attention branch to enhance the network's representational power. The attention branch is gated using the Sigmoid function and multiplied by the feature map's trunk branch. It is too sensitive to coordinate and balance the trunk and attention branches' contributions. To control the attention branch's influence, we propose a new attention method, called Shift-and-Balance (SB). Different from Squeeze-and-Excite, the attention branch is regulated by the learned control factor to control the balance, then added into the feature map's trunk branch. Experiments show that Shift-and-Balance attention significantly improves the accuracy compared to Squeeze-and-Excite when applied in more layers, increasing more size and capacity of a network. Moreover, Shift-and-Balance attention achieves better or close accuracy compared to the state-of-art Dynamic Convolution.

Chunjie Luo, Jianfeng Zhan, Lei Wang et al.

To build light-weight network, we propose a new normalization, Fine-grained Batch Normalization (FBN). Different from Batch Normalization (BN), which normalizes the final summation of the weighted inputs, FBN normalizes the intermediate state of the summation. We propose a novel light-weight network based on FBN, called Finet. At training time, the convolutional layer with FBN can be seen as an inverted bottleneck mechanism. FBN can be fused into convolution at inference time. After fusion, Finet uses the standard convolution with equal channel width, thus makes the inference more efficient. On ImageNet classification dataset, Finet achieves the state-of-art performance (65.706% accuracy with 43M FLOPs, and 73.786% accuracy with 303M FLOPs), Moreover, experiments show that Finet is more efficient than other state-of-art light-weight networks.

Chunjie Luo, Xiwen He, Jianfeng Zhan et al.

Due to increasing amounts of data and compute resources, deep learning achieves many successes in various domains. The application of deep learning on the mobile and embedded devices is taken more and more attentions, benchmarking and ranking the AI abilities of mobile and embedded devices becomes an urgent problem to be solved. Considering the model diversity and framework diversity, we propose a benchmark suite, AIoTBench, which focuses on the evaluation of the inference abilities of mobile and embedded devices. AIoTBench covers three typical heavy-weight networks: ResNet50, InceptionV3, DenseNet121, as well as three light-weight networks: SqueezeNet, MobileNetV2, MnasNet. Each network is implemented by three frameworks which are designed for mobile and embedded devices: Tensorflow Lite, Caffe2, Pytorch Mobile. To compare and rank the AI capabilities of the devices, we propose two unified metrics as the AI scores: Valid Images Per Second (VIPS) and Valid FLOPs Per Second (VOPS). Currently, we have compared and ranked 5 mobile devices using our benchmark. This list will be extended and updated soon after.

Wanling Gao, Fei Tang, Jianfeng Zhan et al.

Modern real-world application scenarios like Internet services consist of a diversity of AI and non-AI modules with huge code sizes and long and complicated execution paths, which raises serious benchmarking or evaluating challenges. Using AI components or micro benchmarks alone can lead to error-prone conclusions. This paper presents a methodology to attack the above challenge. We formalize a real-world application scenario as a Directed Acyclic Graph-based model and propose the rules to distill it into a permutation of essential AI and non-AI tasks, which we call a scenario benchmark. Together with seventeen industry partners, we extract nine typical scenario benchmarks. We design and implement an extensible, configurable, and flexible benchmark framework. We implement two Internet service AI scenario benchmarks based on the framework as proxies to two real-world application scenarios. We consider scenario, component, and micro benchmarks as three indispensable parts for evaluating. Our evaluation shows the advantage of our methodology against using component or micro AI benchmarks alone. The specifications, source code, testbed, and results are publicly available from \url{https://www.benchcouncil.org/aibench/scenario/}.

Fei Tang, Wanling Gao, Jianfeng Zhan et al.

Earlier-stage evaluations of a new AI architecture/system need affordable benchmarks. Only using a few AI component benchmarks like MLPerfalone in the other stages may lead to misleading conclusions. Moreover, the learning dynamics are not well understood, and the benchmarks' shelf-life is short. This paper proposes a balanced benchmarking methodology. We use real-world benchmarks to cover the factors space that impacts the learning dynamics to the most considerable extent. After performing an exhaustive survey on Internet service AI domains, we identify and implement nineteen representative AI tasks with state-of-the-art models. For repeatable performance ranking (RPR subset) and workload characterization (WC subset), we keep two subsets to a minimum for affordability. We contribute by far the most comprehensive AI training benchmark suite. The evaluations show: (1) AIBench Training (v1.1) outperforms MLPerfTraining (v0.7) in terms of diversity and representativeness of model complexity, computational cost, convergent rate, computation, and memory access patterns, and hotspot functions; (2) Against the AIBench full benchmarks, its RPR subset shortens the benchmarking cost by 64%, while maintaining the primary workload characteristics; (3) The performance ranking shows the single-purpose AI accelerator like TPU with the optimized TensorFlowframework performs better than that of GPUs while losing the latter's general support for various AI models. The specification, source code, and performance numbers are available from the AIBench homepage https://www.benchcouncil.org/aibench-training/index.html.

Chunjie Luo, Jianfeng Zhan, Lei Wang et al.

Batch normalization (BN) has become a standard technique for training the modern deep networks. However, its effectiveness diminishes when the batch size becomes smaller, since the batch statistics estimation becomes inaccurate. That hinders batch normalization's usage for 1) training larger model which requires small batches constrained by memory consumption, 2) training on mobile or embedded devices of which the memory resource is limited. In this paper, we propose a simple but effective method, called extended batch normalization (EBN). For NCHW format feature maps, extended batch normalization computes the mean along the (N, H, W) dimensions, as the same as batch normalization, to maintain the advantage of batch normalization. To alleviate the problem caused by small batch size, extended batch normalization computes the standard deviation along the (N, C, H, W) dimensions, thus enlarges the number of samples from which the standard deviation is computed. We compare extended batch normalization with batch normalization and group normalization on the datasets of MNIST, CIFAR-10/100, STL-10, and ImageNet, respectively. The experiments show that extended batch normalization alleviates the problem of batch normalization with small batch size while achieving close performances to batch normalization with large batch size.

Wanling Gao, Fei Tang, Jianfeng Zhan et al.

Domain-specific software and hardware co-design is encouraging as it is much easier to achieve efficiency for fewer tasks. Agile domain-specific benchmarking speeds up the process as it provides not only relevant design inputs but also relevant metrics, and tools. Unfortunately, modern workloads like Big data, AI, and Internet services dwarf the traditional one in terms of code size, deployment scale, and execution path, and hence raise serious benchmarking challenges. This paper proposes an agile domain-specific benchmarking methodology. Together with seventeen industry partners, we identify ten important end-to-end application scenarios, among which sixteen representative AI tasks are distilled as the AI component benchmarks. We propose the permutations of essential AI and non-AI component benchmarks as end-to-end benchmarks. An end-to-end benchmark is a distillation of the essential attributes of an industry-scale application. We design and implement a highly extensible, configurable, and flexible benchmark framework, on the basis of which, we propose the guideline for building end-to-end benchmarks, and present the first end-to-end Internet service AI benchmark. The preliminary evaluation shows the value of our benchmark suite---AIBench against MLPerf and TailBench for hardware and software designers, micro-architectural researchers, and code developers. The specifications, source code, testbed, and results are publicly available from the web site \url{http://www.benchcouncil.org/AIBench/index.html}.

Jianfeng Zhan, Lei Wang, Wanling Gao et al.

This paper outlines BenchCouncil's view on the challenges, rules, and vision of benchmarking modern workloads like Big Data, AI or machine learning, and Internet Services. We conclude the challenges of benchmarking modern workloads as FIDSS (Fragmented, Isolated, Dynamic, Service-based, and Stochastic), and propose the PRDAERS benchmarking rules that the benchmarks should be specified in a paper-and-pencil manner, relevant, diverse, containing different levels of abstractions, specifying the evaluation metrics and methodology, repeatable, and scaleable. We believe proposing simple but elegant abstractions that help achieve both efficiency and general-purpose is the final target of benchmarking in future, which may be not pressing. In the light of this vision, we shortly discuss BenchCouncil's related projects.

Wanling Gao, Fei Tang, Lei Wang et al.

Today's Internet Services are undergoing fundamental changes and shifting to an intelligent computing era where AI is widely employed to augment services. In this context, many innovative AI algorithms, systems, and architectures are proposed, and thus the importance of benchmarking and evaluating them rises. However, modern Internet services adopt a microservice-based architecture and consist of various modules. The diversity of these modules and complexity of execution paths, the massive scale and complex hierarchy of datacenter infrastructure, the confidential issues of data sets and workloads pose great challenges to benchmarking. In this paper, we present the first industry-standard Internet service AI benchmark suite---AIBench with seventeen industry partners, including several top Internet service providers. AIBench provides a highly extensible, configurable, and flexible benchmark framework that contains loosely coupled modules. We identify sixteen prominent AI problem domains like learning to rank, each of which forms an AI component benchmark, from three most important Internet service domains: search engine, social network, and e-commerce, which is by far the most comprehensive AI benchmarking effort. On the basis of the AIBench framework, abstracting the real-world data sets and workloads from one of the top e-commerce providers, we design and implement the first end-to-end Internet service AI benchmark, which contains the primary modules in the critical paths of an industry scale application and is scalable to deploy on different cluster scales. The specifications, source code, and performance numbers are publicly available from the benchmark council web site http://www.benchcouncil.org/AIBench/index.html.

Yunyou Huang, Nana Wang, Wanling Gao et al.

Accurate day-ahead individual residential load forecasting is of great importance to various applications of smart grid on day-ahead market. Deep learning, as a powerful machine learning technology, has shown great advantages and promising application in load forecasting tasks. However, deep learning is a computationally-hungry method, and requires high costs (e.g., time, energy and CO2 emission) to train a deep learning model, which aggravates the energy crisis and incurs a substantial burden to the environment. As a consequence, the deep learning methods are difficult to be popularized and applied in the real smart grid environment. In this paper, we propose a low training cost model based on convolutional neural network, namely LoadCNN, for next-day load forecasting of individual resident with reduced training cost. The experiments show that the training time of LoadCNN is only approximately 1/54 of the one of other state-of-the-art models, and energy consumption and CO2 emissions are only approximate 1/45 of those of other state-of-the-art models based on the same indicators. Meanwhile, the prediction accuracy of our model is equal to that of current state-of-the-art models, making LoadCNN the first load forecasting model simultaneously achieving high prediction accuracy and low training costs. LoadCNN is an efficient green model that is able to be quickly, cost-effectively and environmentally-friendly deployed in a realistic smart grid environment.

Fanda Fan, Yunyou Huang, Lei Wang et al.

It is necessary for clinicians to comprehensively analyze patient information from different sources. Medical image fusion is a promising approach to providing overall information from medical images of different modalities. However, existing medical image fusion approaches ignore the semantics of images, making the fused image difficult to understand. In this work, we propose a new evaluation index to measure the semantic loss of fused image, and put forward a Fusion W-Net (FW-Net) for multimodal medical image fusion. The experimental results are promising: the fused image generated by our approach greatly reduces the semantic information loss, and has better visual effects in contrast to five state-of-art approaches. Our approach and tool have great potential to be applied in the clinical setting.

Yunyou Huang, Zhifei Zhang, Nana Wang et al.

Artificial intelligence (AI) researchers claim that they have made great `achievements' in clinical realms. However, clinicians point out the so-called `achievements' have no ability to implement into natural clinical settings. The root cause for this huge gap is that many essential features of natural clinical tasks are overlooked by AI system developers without medical background. In this paper, we propose that the clinical benchmark suite is a novel and promising direction to capture the essential features of the real-world clinical tasks, hence qualifies itself for guiding the development of AI systems, promoting the implementation of AI in real-world clinical practice.

Wanling Gao, Jianfeng Zhan, Lei Wang et al.

Several fundamental changes in technology indicate domain-specific hardware and software co-design is the only path left. In this context, architecture, system, data management, and machine learning communities pay greater attention to innovative big data and AI algorithms, architecture, and systems. Unfortunately, complexity, diversity, frequently-changed workloads, and rapid evolution of big data and AI systems raise great challenges. First, the traditional benchmarking methodology that creates a new benchmark or proxy for every possible workload is not scalable, or even impossible for Big Data and AI benchmarking. Second, it is prohibitively expensive to tailor the architecture to characteristics of one or more application or even a domain of applications. We consider each big data and AI workload as a pipeline of one or more classes of units of computation performed on different initial or intermediate data inputs, each class of which we call a data motif. On the basis of our previous work that identifies eight data motifs taking up most of the run time of a wide variety of big data and AI workloads, we propose a scalable benchmarking methodology that uses the combination of one or more data motifs---to represent diversity of big data and AI workloads. Following this methodology, we present a unified big data and AI benchmark suite---BigDataBench 4.0, publicly available from~\url{http://prof.ict.ac.cn/BigDataBench}. This unified benchmark suite sheds new light on domain-specific hardware and software co-design: tailoring the system and architecture to characteristics of the unified eight data motifs other than one or more application case by case. Also, for the first time, we comprehensively characterize the CPU pipeline efficiency using the benchmarks of seven workload types in BigDataBench 4.0.

Jiechao Cheng, Rui Ren, Lei Wang et al.

The increasing popularity of server usage has brought a plenty of anomaly log events, which have threatened a vast collection of machines. Recognizing and categorizing the anomalous events thereby is a much salient work for our systems, especially the ones generate the massive amount of data and harness it for technology value creation and business development. To assist in focusing on the classification and the prediction of anomaly events, and gaining critical insights from system event records, we propose a novel log preprocessing method which is very effective to filter abundant information and retain critical characteristics. Additionally, a competitive approach for automated classification of anomalous events detected from the distributed system logs with the state-of-the-art deep (Convolutional Neural Network) architectures is proposed in this paper. We measure a series of deep CNN algorithms with varied hyper-parameter combinations by using standard evaluation metrics, the results of our study reveals the advantages and potential capabilities of the proposed deep CNN models for anomaly event classification tasks on real-world systems. The optimal classification precision of our approach is 98.14%, which surpasses the popular traditional machine learning methods.

Chunjie Luo, Jianfeng Zhan, Lei Wang et al.

Traditionally, multi-layer neural networks use dot product between the output vector of previous layer and the incoming weight vector as the input to activation function. The result of dot product is unbounded, thus increases the risk of large variance. Large variance of neuron makes the model sensitive to the change of input distribution, thus results in poor generalization, and aggravates the internal covariate shift which slows down the training. To bound dot product and decrease the variance, we propose to use cosine similarity or centered cosine similarity (Pearson Correlation Coefficient) instead of dot product in neural networks, which we call cosine normalization. We compare cosine normalization with batch, weight and layer normalization in fully-connected neural networks as well as convolutional networks on the data sets of MNIST, 20NEWS GROUP, CIFAR-10/100 and SVHN. Experiments show that cosine normalization achieves better performance than other normalization techniques.