Engineering on 中文AI术语词典https://terms-en.ai-term-hub.com/zh/tags/engineering/Recent content in Engineering on 中文AI术语词典Hugozh-cnSat, 18 Jul 2026 11:44:45 +0000吞吐量https://terms-en.ai-term-hub.com/zh/terms/throughput/Sat, 18 Jul 2026 11:36:49 +0000https://terms-en.ai-term-hub.com/zh/terms/throughput/<h2 id="definition">Definition</h2> <p>在人工智能工程中，吞吐量是一个关键的性能指标，用于指示系统容量。对于大语言模型（LLMs），它通常以每秒令牌数（tokens per second）来衡量；对于计算机视觉模型，则以每秒图像数来衡量；对于查询处理，则可能以每秒查询数来表示。高吞吐量意味着系统能够高效地处理大量并发任务。</p> <h3 id="summary">Summary</h3> <p>吞吐量衡量人工智能系统在给定时间范围内成功处理的数据量或请求数量。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>性能指标</li> <li>可扩展性</li> <li>批处理</li> <li>延迟与吞吐量</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>大语言模型推理服务</li> <li>实时视频处理</li> <li>高并发API设计</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/latency-%E5%BB%B6%E8%BF%9F/">Latency (延迟)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/batch-size-%E6%89%B9%E5%A4%A7%E5%B0%8F/">Batch Size (批大小)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/gpu-utilization-gpu%E5%88%A9%E7%94%A8%E7%8E%87/">GPU Utilization (GPU利用率)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/model-serving-%E6%A8%A1%E5%9E%8B%E6%9C%8D%E5%8A%A1/">Model Serving (模型服务)</a></li> </ul>SentencePiecehttps://terms-en.ai-term-hub.com/zh/terms/sentencepiece/Sat, 18 Jul 2026 11:33:07 +0000https://terms-en.ai-term-hub.com/zh/terms/sentencepiece/<h2 id="definition">Definition</h2> <p>SentencePiece 是一个流行的开源文本归一化和分词库，广泛用于现代自然语言处理管道中。它执行联合词块和子词词汇表的无监督学习。</p> <h3 id="summary">Summary</h3> <p>一种无监督文本分词器和去分词器库，将原始文本视为子词序列用于自然语言处理预处理。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>子词分词</li> <li>词汇表学习</li> <li>去分词</li> <li>语言无关</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>为变换器模型预处理数据</li> <li>处理多语言文本语料库</li> <li>减少语言模型的词汇量大小</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/tokenizer-%E5%88%86%E8%AF%8D%E5%99%A8/">Tokenizer (分词器)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bpe-%E5%AD%97%E8%8A%82%E5%AF%B9%E7%BC%96%E7%A0%81/">BPE (字节对编码)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/byte-pair-encoding-%E5%AD%97%E8%8A%82%E5%AF%B9%E7%BC%96%E7%A0%81/">Byte-Pair Encoding (字节对编码)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/nlp-preprocessing-%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E5%A4%84%E7%90%86%E9%A2%84%E5%A4%84%E7%90%86/">NLP Preprocessing (自然语言处理预处理)</a></li> </ul>可观测性https://terms-en.ai-term-hub.com/zh/terms/observability/Sat, 18 Jul 2026 11:28:49 +0000https://terms-en.ai-term-hub.com/zh/terms/observability/<h2 id="definition">Definition</h2> <p>在人工智能工程中，可观测性指的是通过分析机器学习系统的外部输出来理解其内部状态的能力。它超越了传统的监控，旨在提供对复杂系统运行状况的深入洞察。</p> <h3 id="summary">Summary</h3> <p>可观测性是衡量能否通过日志、指标和追踪等外部输出来推断系统内部状态程度的指标。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>指标</li> <li>日志</li> <li>追踪</li> <li>根本原因分析</li> <li>系统状态推断</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>调试生产环境中的机器学习流水线</li> <li>监控模型性能漂移</li> <li>优化推理延迟</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/monitoring-%E7%9B%91%E6%8E%A7/">monitoring (监控)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/telemetry-%E9%81%A5%E6%B5%8B/">telemetry (遥测)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/distributed_tracing-%E5%88%86%E5%B8%83%E5%BC%8F%E8%BF%BD%E8%B8%AA/">distributed_tracing (分布式追踪)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/ml_ops-%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E8%BF%90%E7%BB%B4/">ml_ops (机器学习运维)</a></li> </ul>模型注册表https://terms-en.ai-term-hub.com/zh/terms/model_registry/Sat, 18 Jul 2026 11:26:37 +0000https://terms-en.ai-term-hub.com/zh/terms/model_registry/<h2 id="definition">Definition</h2> <p>模型注册表是 MLOps 中的关键组件，提供用于存储、版本控制和管理的统一仓库。它使团队能够跟踪模型元数据、性能指标等。</p> <h3 id="summary">Summary</h3> <p>用于在整个机器学习生命周期中跟踪和管理模型的集中式存储和系统。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>版本控制</li> <li>元数据管理</li> <li>生命周期跟踪</li> <li>MLOps</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>跟踪从开发到生产的模型版本</li> <li>审计模型性能和血缘关系</li> <li>促进团队在模型工件上的协作</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mlflow-%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E6%B5%81%E6%B0%B4%E7%BA%BF%E5%B9%B3%E5%8F%B0/">MLflow (机器学习流水线平台)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2-model-deployment/">模型部署 (Model Deployment)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%89%B9%E5%BE%81%E5%AD%98%E5%82%A8-feature-store/">特征存储 (Feature Store)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%8C%81%E7%BB%AD%E9%9B%86%E6%88%90-%E6%8C%81%E7%BB%AD%E4%BA%A4%E4%BB%98-ci-cd/">持续集成/持续交付 (CI/CD)</a></li> </ul>基于知识的配置https://terms-en.ai-term-hub.com/zh/terms/knowledge_based_configuration/Sat, 18 Jul 2026 11:23:17 +0000https://terms-en.ai-term-hub.com/zh/terms/knowledge_based_configuration/<h2 id="definition">Definition</h2> <p>该方法在知识库中采用约束满足技术，以确保组装的产品符合所有技术和客户要求。它通过编码规则来防止无效的组合。</p> <h3 id="summary">Summary</h3> <p>一种自动化过程，利用领域特定的知识库，根据用户约束生成有效的产品配置。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>约束满足</li> <li>产品建模</li> <li>基于规则的系統</li> <li>自动推理</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>定制计算机硬件组装</li> <li>软件许可证和功能选择</li> <li>工业机械设置</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/cp-sat-%E7%BA%A6%E6%9D%9F%E8%A7%84%E5%88%92%E6%B1%82%E8%A7%A3%E5%99%A8/">CP-SAT (约束规划求解器)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/expert-systems-%E4%B8%93%E5%AE%B6%E7%B3%BB%E7%BB%9F/">Expert Systems (专家系统)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/product-lifecycle-management-%E4%BA%A7%E5%93%81%E7%94%9F%E5%91%BD%E5%91%A8%E6%9C%9F%E7%AE%A1%E7%90%86/">Product Lifecycle Management (产品生命周期管理)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/constraint-programming-%E7%BA%A6%E6%9D%9F%E7%BC%96%E7%A8%8B/">Constraint Programming (约束编程)</a></li> </ul>智能控制https://terms-en.ai-term-hub.com/zh/terms/intelligent_control/Sat, 18 Jul 2026 11:22:32 +0000https://terms-en.ai-term-hub.com/zh/terms/intelligent_control/<h2 id="definition">Definition</h2> <p>智能控制采用模糊逻辑、神经网络和遗传算法等人工智能方法来调节那些传统数学建模不足或</p> <h3 id="summary">Summary</h3> <p>利用AI技术来管理复杂、非线性或不确定动态过程的控制系统的统称。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>自适应控制</li> <li>非线性系统</li> <li>模糊逻辑</li> <li>神经网络</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>湍流风况下的无人机自主稳定</li> <li>智能电网能源分配优化</li> <li>工业机器人手臂的精密操作</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/feedback_control-%E5%8F%8D%E9%A6%88%E6%8E%A7%E5%88%B6/">feedback_control (反馈控制)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/adaptive_systems-%E8%87%AA%E9%80%82%E5%BA%94%E7%B3%BB%E7%BB%9F/">adaptive_systems (自适应系统)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/robotics-%E6%9C%BA%E5%99%A8%E4%BA%BA%E5%AD%A6/">robotics (机器人学)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/fuzzy_logic-%E6%A8%A1%E7%B3%8A%E9%80%BB%E8%BE%91/">fuzzy_logic (模糊逻辑)</a></li> </ul>GraphQLhttps://terms-en.ai-term-hub.com/zh/terms/graphql/Sat, 18 Jul 2026 11:19:45 +0000https://terms-en.ai-term-hub.com/zh/terms/graphql/<h2 id="definition">Definition</h2> <p>由 Facebook 开发，GraphQL 提供了对 API 中数据的完整且易于理解的描述，赋予客户端只请求所需数据的能力。它取代了传统的 REST API 模式，减少了过度获取或获取不足的问题。</p> <h3 id="summary">Summary</h3> <p>GraphQL 是一种用于 API 的查询语言，以及根据现有数据满足这些查询的运行时，允许客户端请求它们确切需要的内容。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>模式定义</li> <li>查询语言</li> <li>单一端点</li> <li>类型安全</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>现代 Web 应用后端</li> <li>移动应用数据获取</li> <li>微服务编排</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/rest-api/">REST API</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/apollo-server/">Apollo Server</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/json/">JSON</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%A8%A1%E5%BC%8F/">模式</a></li> </ul>前瞻技术https://terms-en.ai-term-hub.com/zh/terms/forethought_technologies/Sat, 18 Jul 2026 11:17:40 +0000https://terms-en.ai-term-hub.com/zh/terms/forethought_technologies/<h2 id="definition">Definition</h2> <p>这一概念涉及设计具有前瞻性能力的AI系统，使其能够模拟潜在结果并主动适应变化。它整合了预测分析、情景规划和风险管理，旨在提高系统的鲁棒性和决策质量，从而在问题发生前进行干预。</p> <h3 id="summary">Summary</h3> <p>前瞻技术是指通过预测建模和模拟来预见未来系统状态、风险和需求的工程实践。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>预测建模</li> <li>风险评估</li> <li>情景规划</li> <li>主动设计</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>自动驾驶导航</li> <li>金融欺诈检测</li> <li>基础设施维护预测</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/predictive_analytics-%E9%A2%84%E6%B5%8B%E5%88%86%E6%9E%90/">predictive_analytics (预测分析)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/simulation-%E6%A8%A1%E6%8B%9F/">simulation (模拟)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/risk_management-%E9%A3%8E%E9%99%A9%E7%AE%A1%E7%90%86/">risk_management (风险管理)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/planning-%E8%A7%84%E5%88%92/">planning (规划)</a></li> </ul>实验跟踪https://terms-en.ai-term-hub.com/zh/terms/experiment_tracking/Sat, 18 Jul 2026 11:16:49 +0000https://terms-en.ai-term-hub.com/zh/terms/experiment_tracking/<h2 id="definition">Definition</h2> <p>这种做法涉及在训练过程中记录超参数、数据集版本、模型架构和性能指标。它使数据科学家能够比较不同的实验配置。</p> <h3 id="summary">Summary</h3> <p>实验跟踪是系统记录机器学习实验的元数据、指标和工件的过程，以确保可复现性并促进比较。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>可复现性</li> <li>超参数记录</li> <li>工件管理</li> <li>版本控制</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>比较不同超参数设置下的模型性能</li> <li>通过查看记录的指标来调试训练故障</li> <li>与团队成员协作进行共享实验</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mlflow/">MLflow</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%A8%A1%E5%9E%8B%E6%B3%A8%E5%86%8C%E8%A1%A8/">模型注册表</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E8%B6%85%E5%8F%82%E6%95%B0%E8%B0%83%E4%BC%98/">超参数调优</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%95%B0%E6%8D%AE%E7%89%88%E6%9C%AC%E6%8E%A7%E5%88%B6/">数据版本控制</a></li> </ul>Eaglehttps://terms-en.ai-term-hub.com/zh/terms/eagle/Sat, 18 Jul 2026 11:15:46 +0000https://terms-en.ai-term-hub.com/zh/terms/eagle/<h2 id="definition">Definition</h2> <p>Eagle代表了在大语言模型领域内的一种特定架构和工程框架，主要侧重于训练效率和可扩展性的优化。它通过改进底层实现和分布式策略，帮助研究人员和企业更快速地训练和部署大规模语言模型。</p> <h3 id="summary">Summary</h3> <p>Eagle是由字节跳动开发的高性能开源大语言模型框架，旨在高效地进行大模型的预训练和微调。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>大语言模型框架</li> <li>训练效率</li> <li>Transformer架构</li> <li>开源</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>大规模模型预训练</li> <li>大语言模型的高效微调</li> <li>模型优化研究</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/pytorch-%E6%B5%81%E8%A1%8C%E7%9A%84%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E6%A1%86%E6%9E%B6/">PyTorch (流行的深度学习框架)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/megatron-lm-nvidia%E5%BC%80%E5%8F%91%E7%9A%84%E5%A4%A7%E6%A8%A1%E5%9E%8B%E8%AE%AD%E7%BB%83%E6%A1%86%E6%9E%B6/">Megatron-LM (NVIDIA开发的大模型训练框架)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/deepspeed-%E5%BE%AE%E8%BD%AF%E5%BC%80%E5%8F%91%E7%9A%84%E5%A4%A7%E8%A7%84%E6%A8%A1%E6%A8%A1%E5%9E%8B%E8%AE%AD%E7%BB%83%E5%BA%93/">DeepSpeed (微软开发的大规模模型训练库)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%A4%A7%E8%AF%AD%E8%A8%80%E6%A8%A1%E5%9E%8B-%E5%85%B7%E6%9C%89%E6%95%B0%E5%8D%81%E4%BA%BF%E5%8F%82%E6%95%B0%E4%BB%A5%E4%B8%8A%E7%9A%84%E8%AF%AD%E8%A8%80%E6%A8%A1%E5%9E%8B/">大语言模型 (具有数十亿参数以上的语言模型)</a></li> </ul>裁剪https://terms-en.ai-term-hub.com/zh/terms/clip/Sat, 18 Jul 2026 11:10:07 +0000https://terms-en.ai-term-hub.com/zh/terms/clip/<h2 id="definition">Definition</h2> <p>在深度学习工程中，裁剪通常应用于梯度以缓解梯度爆炸问题，确保反向传播的稳定性。它也可以指在应用激活函数之前限制输出 logits 的范围。</p> <h3 id="summary">Summary</h3> <p>裁剪是一种限制数值（如梯度或输出概率）幅度的技术，旨在防止训练过程中的数值不稳定。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>梯度裁剪</li> <li>数值稳定性</li> <li>梯度爆炸</li> <li>正则化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>循环神经网络训练</li> <li>稳定 Transformer 训练</li> <li>防止损失发散</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/learning-rate-%E5%AD%A6%E4%B9%A0%E7%8E%87/">Learning Rate (学习率)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/backpropagation-%E5%8F%8D%E5%90%91%E4%BC%A0%E6%92%AD/">Backpropagation (反向传播)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/vanishing-gradient-%E6%A2%AF%E5%BA%A6%E6%B6%88%E5%A4%B1/">Vanishing Gradient (梯度消失)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/normalization-%E5%BD%92%E4%B8%80%E5%8C%96/">Normalization (归一化)</a></li> </ul>缓存https://terms-en.ai-term-hub.com/zh/terms/caching/Sat, 18 Jul 2026 11:09:42 +0000https://terms-en.ai-term-hub.com/zh/terms/caching/<h2 id="definition">Definition</h2> <p>在AI工程中，缓存通过将最近或频繁的查询结果、模型预测或中间计算保留在快速内存（如RAM）中来优化性能。这减少了昂贵的主数据存储访问需求，从而显著提升系统响应速度。</p> <h3 id="summary">Summary</h3> <p>缓存是一种将频繁访问的数据存储在临时高速存储层中的技术，旨在降低延迟并减少主数据源的负载。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>延迟降低</li> <li>内存优化</li> <li>驱逐策略</li> <li>命中率</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>存储模型推理结果</li> <li>缓存数据库查询输出</li> <li>预计算特征嵌入</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">import</span> redis </span></span><span style="display:flex;"><span> </span></span><span style="display:flex;"><span><span style="color:#75715e"># Simple caching example</span> </span></span><span style="display:flex;"><span>r <span style="color:#f92672">=</span> redis<span style="color:#f92672">.</span>Redis(host<span style="color:#f92672">=</span><span style="color:#e6db74">&#39;localhost&#39;</span>, port<span style="color:#f92672">=</span><span style="color:#ae81ff">6379</span>, db<span style="color:#f92672">=</span><span style="color:#ae81ff">0</span>) </span></span><span style="display:flex;"><span> </span></span><span style="display:flex;"><span><span style="color:#66d9ef">def</span> <span style="color:#a6e22e">get_prediction</span>(model_id, input_data): </span></span><span style="display:flex;"><span> cache_key <span style="color:#f92672">=</span> <span style="color:#e6db74">f</span><span style="color:#e6db74">&#34;pred_</span><span style="color:#e6db74">{</span>model_id<span style="color:#e6db74">}</span><span style="color:#e6db74">_</span><span style="color:#e6db74">{</span>hash(str(input_data))<span style="color:#e6db74">}</span><span style="color:#e6db74">&#34;</span> </span></span><span style="display:flex;"><span> result <span style="color:#f92672">=</span> r<span style="color:#f92672">.</span>get(cache_key) </span></span><span style="display:flex;"><span> <span style="color:#66d9ef">if</span> result: </span></span><span style="display:flex;"><span> <span style="color:#66d9ef">return</span> result<span style="color:#f92672">.</span>decode(<span style="color:#e6db74">&#39;utf-8&#39;</span>) </span></span><span style="display:flex;"><span> <span style="color:#75715e"># Compute if not cached</span> </span></span><span style="display:flex;"><span> prediction <span style="color:#f92672">=</span> model<span style="color:#f92672">.</span>predict(input_data) </span></span><span style="display:flex;"><span> r<span style="color:#f92672">.</span>setex(cache_key, <span style="color:#ae81ff">3600</span>, str(prediction)) </span></span><span style="display:flex;"><span> <span style="color:#66d9ef">return</span> prediction </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/redis-%E5%86%85%E5%AD%98%E6%95%B0%E6%8D%AE%E7%BB%93%E6%9E%84%E5%AD%98%E5%82%A8/">Redis (内存数据结构存储)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/memcached-%E5%88%86%E5%B8%83%E5%BC%8F%E5%86%85%E5%AD%98%E7%BC%93%E5%AD%98%E7%B3%BB%E7%BB%9F/">memcached (分布式内存缓存系统)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/performance-tuning-%E6%80%A7%E8%83%BD%E8%B0%83%E4%BC%98/">performance tuning (性能调优)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/database-indexing-%E6%95%B0%E6%8D%AE%E5%BA%93%E7%B4%A2%E5%BC%95/">database indexing (数据库索引)</a></li> </ul>论证框架https://terms-en.ai-term-hub.com/zh/terms/argumentation_framework/Sat, 18 Jul 2026 11:05:01 +0000https://terms-en.ai-term-hub.com/zh/terms/argumentation_framework/<h2 id="definition">Definition</h2> <p>论证框架为表示论点及其之间的攻击和防御提供了数学基础。在AI工程中，它们帮助系统通过权衡证据做出透明且可辩护的决策。</p> <h3 id="summary">Summary</h3> <p>用于建模和解决AI系统中竞争性主张或决策之间冲突的形式化结构。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>冲突解决</li> <li>可解释AI</li> <li>逻辑推理</li> <li>决策透明度</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>法律推理系统</li> <li>医学诊断依据说明</li> <li>内容审核政策</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%8F%AF%E8%A7%A3%E9%87%8Aai-xai/">可解释AI (XAI)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E9%80%BB%E8%BE%91%E7%BC%96%E7%A8%8B-logic-programming/">逻辑编程 (Logic Programming)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%86%B3%E7%AD%96%E6%94%AF%E6%8C%81%E7%B3%BB%E7%BB%9F-decision-support-systems/">决策支持系统 (Decision Support Systems)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%9F%A5%E8%AF%86%E5%9B%BE%E8%B0%B1-knowledge-graphs/">知识图谱 (Knowledge Graphs)</a></li> </ul>算法选择https://terms-en.ai-term-hub.com/zh/terms/algorithm_selection/Sat, 18 Jul 2026 11:04:37 +0000https://terms-en.ai-term-hub.com/zh/terms/algorithm_selection/<h2 id="definition">Definition</h2> <p>算法选择涉及评估不同的计算方法，以确定哪一种能最高效地解决给定任务。此过程考虑的因素包括时间复杂度、空间复杂</p> <h3 id="summary">Summary</h3> <p>算法选择是根据性能指标和约束条件，为特定问题选择最合适的计算方法的过程。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>性能指标</li> <li>复杂度分析</li> <li>基准测试</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>为大型数据集选择排序算法</li> <li>为机器学习流水线选择分类器</li> <li>优化数据库查询执行计划</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/big-o-notation-%E5%A4%A7o%E7%AC%A6%E5%8F%B7/">Big O Notation (大O符号)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/hyperparameter-tuning-%E8%B6%85%E5%8F%82%E6%95%B0%E8%B0%83%E4%BC%98/">Hyperparameter Tuning (超参数调优)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/model-evaluation-%E6%A8%A1%E5%9E%8B%E8%AF%84%E4%BC%B0/">Model Evaluation (模型评估)</a></li> </ul>AI基础设施https://terms-en.ai-term-hub.com/zh/terms/ai_infrastructure/Sat, 18 Jul 2026 11:03:04 +0000https://terms-en.ai-term-hub.com/zh/terms/ai_infrastructure/<h2 id="definition">Definition</h2> <p>AI基础设施涵盖了人工智能运营所需的基础技术栈。这包括高性能计算硬件（如GPU和TPU）、云存储解决方案以及支持数据流动的网络架构，旨在确保AI系统的高效运行和扩展能力。</p> <h3 id="summary">Summary</h3> <p>用于大规模开发、训练和部署人工智能模型所需的硬件、软件和网络资源。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>GPU/TPU集群</li> <li>云计算</li> <li>数据管道</li> <li>模型部署</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>规划大规模模型训练环境</li> <li>优化AI工作负载的云成本</li> <li>为生产系统构建MLOps管道</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mlops-%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E8%BF%90%E7%BB%B4/">MLOps (机器学习运维)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E4%BA%91%E8%AE%A1%E7%AE%97-cloud-computing/">云计算 (Cloud Computing)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E9%AB%98%E6%80%A7%E8%83%BD%E8%AE%A1%E7%AE%97-high-performance-computing/">高性能计算 (High-Performance Computing)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%95%B0%E6%8D%AE%E5%B7%A5%E7%A8%8B-data-engineering/">数据工程 (Data Engineering)</a></li> </ul>AI可观测性https://terms-en.ai-term-hub.com/zh/terms/ai_observability/Sat, 18 Jul 2026 11:03:04 +0000https://terms-en.ai-term-hub.com/zh/terms/ai_observability/<h2 id="definition">Definition</h2> <p>AI可观测性将传统的软件监控扩展到解决机器学习系统的独特挑战。它涉及实时跟踪模型性能、数据漂移和推理延迟，从而帮助开发者深入理解黑盒模型的决策过程及其在生产环境中的行为。</p> <h3 id="summary">Summary</h3> <p>通过日志、指标和追踪来监控和理解机器学习系统内部状态的做法。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>数据漂移</li> <li>模型监控</li> <li>遥测</li> <li>调试</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>检测生产模型中的概念漂移</li> <li>排查低置信度预测问题</li> <li>确保AI服务符合服务等级协议(SLA)</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">import</span> mlflow </span></span><span style="display:flex;"><span> </span></span><span style="display:flex;"><span><span style="color:#75715e"># Log metrics during training</span> </span></span><span style="display:flex;"><span>mlflow<span style="color:#f92672">.</span>log_metric(<span style="color:#e6db74">&#39;accuracy&#39;</span>, <span style="color:#ae81ff">0.95</span>) </span></span><span style="display:flex;"><span>mlflow<span style="color:#f92672">.</span>log_metric(<span style="color:#e6db74">&#39;loss&#39;</span>, <span style="color:#ae81ff">0.05</span>) </span></span><span style="display:flex;"><span> </span></span><span style="display:flex;"><span><span style="color:#75715e"># Track model parameters</span> </span></span><span style="display:flex;"><span>mlflow<span style="color:#f92672">.</span>log_param(<span style="color:#e6db74">&#39;learning_rate&#39;</span>, <span style="color:#ae81ff">0.01</span>) </span></span><span style="display:flex;"><span>mlflow<span style="color:#f92672">.</span>log_param(<span style="color:#e6db74">&#39;epochs&#39;</span>, <span style="color:#ae81ff">10</span>) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mlops-%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E8%BF%90%E7%BB%B4/">MLOps (机器学习运维)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%A8%A1%E5%9E%8B%E6%BC%82%E7%A7%BB-model-drift/">模型漂移 (Model Drift)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%B3%BB%E7%BB%9F%E7%9B%91%E6%8E%A7-system-monitoring/">系统监控 (System Monitoring)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E9%81%A5%E6%B5%8B-telemetry/">遥测 (Telemetry)</a></li> </ul>测试https://terms-en.ai-term-hub.com/zh/terms/testing/Sat, 18 Jul 2026 11:02:04 +0000https://terms-en.ai-term-hub.com/zh/terms/testing/<h2 id="definition">Definition</h2> <p>AI 工程中的测试涉及使用多样化的数据集严格评估模型，以识别偏见、错误和鲁棒性问题。它包括对代码组件的单元测试、集成测试等。</p> <h3 id="summary">Summary</h3> <p>系统地评估 AI 模型在未见过数据上的性能和可靠性，以确保质量和安全的过程。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>评估指标</li> <li>偏见检测</li> <li>鲁棒性</li> <li>生产就绪状态</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>部署前验证模型准确性</li> <li>检测对抗性漏洞</li> <li>确保符合伦理准则</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> sklearn.metrics <span style="color:#f92672">import</span> accuracy_score </span></span><span style="display:flex;"><span>y_pred <span style="color:#f92672">=</span> model<span style="color:#f92672">.</span>predict(X_test) </span></span><span style="display:flex;"><span>accuracy <span style="color:#f92672">=</span> accuracy_score(y_test, y_pred) </span></span><span style="display:flex;"><span><span style="color:#66d9ef">assert</span> accuracy <span style="color:#f92672">&gt;</span> <span style="color:#ae81ff">0.9</span>, <span style="color:#e6db74">&#34;Model accuracy below threshold&#34;</span> </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E9%AA%8C%E8%AF%81-validation/">验证 (Validation)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%9F%BA%E5%87%86%E6%B5%8B%E8%AF%95-benchmarking/">基准测试 (Benchmarking)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%8C%81%E7%BB%AD%E9%9B%86%E6%88%90-%E6%8C%81%E7%BB%AD%E4%BA%A4%E4%BB%98-ci-cd/">持续集成/持续交付 (CI/CD)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%A8%A1%E5%9E%8B%E8%AF%84%E4%BC%B0-model-evaluation/">模型评估 (Model Evaluation)</a></li> </ul>软件开发工具包https://terms-en.ai-term-hub.com/zh/terms/sdk/Sat, 18 Jul 2026 11:01:52 +0000https://terms-en.ai-term-hub.com/zh/terms/sdk/<h2 id="definition">Definition</h2> <p>SDK 是一组软件开发工具，允许开发人员为特定平台或服务创建应用程序。对于 AI 而言，SDK 提供预建的库、API 和实用程序来&hellip;</p> <h3 id="summary">Summary</h3> <p>提供用于构建应用程序的工具、库和文档的软件套件。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>API 集成</li> <li>库</li> <li>开发工具</li> <li>抽象</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>将大语言模型集成到应用中</li> <li>构建计算机视觉流水线</li> <li>将机器学习模型部署到生产环境</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span>null </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/api-%E5%BA%94%E7%94%A8%E7%A8%8B%E5%BA%8F%E7%BC%96%E7%A8%8B%E6%8E%A5%E5%8F%A3/">api (应用程序编程接口)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/library-%E5%BA%93/">library (库)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/framework-%E6%A1%86%E6%9E%B6/">framework (框架)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/toolkit-%E5%B7%A5%E5%85%B7%E5%8C%85/">toolkit (工具包)</a></li> </ul>模型上下文协议https://terms-en.ai-term-hub.com/zh/terms/model_context_protocol/Sat, 18 Jul 2026 11:01:05 +0000https://terms-en.ai-term-hub.com/zh/terms/model_context_protocol/<h2 id="definition">Definition</h2> <p>模型上下文协议（MCP）是一种开放标准，使AI应用能够以统一的方式连接各种数据源，如数据库、API和文件系统。它抽象了连接细节，允许开发人员轻松构建可移植的AI助手，并确保不同系统间的数据集成和工具调用具备互操作性和安全性。</p> <h3 id="summary">Summary</h3> <p>旨在促进AI模型与外部数据源或工具之间安全高效通信的标准化框架。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>互操作性</li> <li>标准化</li> <li>数据集成</li> <li>工具调用</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>将大语言模型连接至企业数据</li> <li>构建可移植的AI助手</li> <li>安全的API集成</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/api-%E5%BA%94%E7%94%A8%E7%A8%8B%E5%BA%8F%E7%BC%96%E7%A8%8B%E6%8E%A5%E5%8F%A3/">API (应用程序编程接口)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/integration-%E9%9B%86%E6%88%90/">Integration (集成)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/llm-%E5%A4%A7%E5%9E%8B%E8%AF%AD%E8%A8%80%E6%A8%A1%E5%9E%8B/">LLM (大型语言模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/interoperability-%E4%BA%92%E6%93%8D%E4%BD%9C%E6%80%A7/">Interoperability (互操作性)</a></li> </ul>延迟https://terms-en.ai-term-hub.com/zh/terms/latency/Sat, 18 Jul 2026 11:00:35 +0000https://terms-en.ai-term-hub.com/zh/terms/latency/<h2 id="definition">Definition</h2> <p>延迟衡量AI服务的响应速度，通常以毫秒为单位表示。它包括推理时间、网络传输延迟和处理开销。低延迟对于实时应用至关重要。</p> <h3 id="summary">Summary</h3> <p>AI系统中请求发起与响应开始之间的时间延迟。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>推理时间</li> <li>响应时间</li> <li>实时处理</li> <li>优化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>语音识别系统</li> <li>自动驾驶车辆控制</li> <li>实时翻译服务</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/throughput-%E5%90%9E%E5%90%90%E9%87%8F/">throughput (吞吐量)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/inference-%E6%8E%A8%E7%90%86/">inference (推理)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/optimization-%E4%BC%98%E5%8C%96/">optimization (优化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/real_time-%E5%AE%9E%E6%97%B6/">real_time (实时)</a></li> </ul>权衡https://terms-en.ai-term-hub.com/zh/terms/trade_off/Sat, 18 Jul 2026 10:59:01 +0000https://terms-en.ai-term-hub.com/zh/terms/trade_off/<h2 id="definition">Definition</h2> <p>在人工智能和工程领域，权衡指的是在优化相互冲突的目标（如模型准确率与计算成本，或延迟与精度）时所需的平衡。由于资源有限，必须在不同指标之间做出取舍。</p> <h3 id="summary">Summary</h3> <p>一种妥协，即在某一领域获得优势的同时，在另一领域遭受损失。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>资源分配</li> <li>优化</li> <li>准确率与速度的权衡</li> <li>偏差-方差权衡</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>为边缘设备选择模型大小</li> <li>平衡推理延迟与预测准确率</li> <li>训练期间管理内存使用量</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/optimization-%E4%BC%98%E5%8C%96/">optimization (优化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/hyperparameter_tuning-%E8%B6%85%E5%8F%82%E6%95%B0%E8%B0%83%E4%BC%98/">hyperparameter_tuning (超参数调优)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/model_compression-%E6%A8%A1%E5%9E%8B%E5%8E%8B%E7%BC%A9/">model_compression (模型压缩)</a></li> </ul>机器人（复数）https://terms-en.ai-term-hub.com/zh/terms/robots/Sat, 18 Jul 2026 10:54:27 +0000https://terms-en.ai-term-hub.com/zh/terms/robots/<h2 id="definition">Definition</h2> <p>机器人涵盖了一类多样化的机器，可根据其移动能力、结构或应用领域进行分类。该类别包括工业机械臂、自主移动机器人（AMR）、无人机等。</p> <h3 id="summary">Summary</h3> <p>指代多个旨在自主执行任务的可编程机器的复数形式。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>多智能体系统</li> <li>机器人工程</li> <li>分类</li> <li>部署</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>仓库物流车队</li> <li>蜂群无人机操作</li> <li>教育套件</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/robotics-%E6%9C%BA%E5%99%A8%E4%BA%BA%E5%AD%A6/">Robotics (机器人学)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/swarm-intelligence-%E7%BE%A4%E4%BD%93%E6%99%BA%E8%83%BD/">Swarm Intelligence (群体智能)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/industrial-automation-%E5%B7%A5%E4%B8%9A%E8%87%AA%E5%8A%A8%E5%8C%96/">Industrial Automation (工业自动化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/cobots-%E5%8D%8F%E4%BD%9C%E6%9C%BA%E5%99%A8%E4%BA%BA/">Cobots (协作机器人)</a></li> </ul>构建https://terms-en.ai-term-hub.com/zh/terms/building/Sat, 18 Jul 2026 10:49:30 +0000https://terms-en.ai-term-hub.com/zh/terms/building/<h2 id="definition">Definition</h2> <p>构建指的是创建AI解决方案的端到端工程过程，包括数据收集、模型选择、训练、验证和部署。它涵盖了技术支持的基础设施。</p> <h3 id="summary">Summary</h3> <p>从初始设计到生产就绪阶段，开发、训练和部署AI模型及系统的实际阶段。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>开发生命周期</li> <li>基础设施</li> <li>部署</li> <li>工程</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>为电子商务平台构建推荐引擎。</li> <li>设置用于持续集成和交付的MLOps流水线。</li> <li>将计算机视觉模型集成到移动应用程序中。</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/development-%E5%BC%80%E5%8F%91/">development (开发)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mlops-%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E8%BF%90%E7%BB%B4/">MLOps (机器学习运维)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/deployment-%E9%83%A8%E7%BD%B2/">deployment (部署)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/engineering-%E5%B7%A5%E7%A8%8B/">engineering (工程)</a></li> </ul>