- Published on
年度Review前夜的深夜独白。关于焦虑、孤独、VibeCoding带来的身份危机，以及在AI革命浪潮中一个普通程序员的迷茫与思考。
All Posts
All Posts
- machine-learning (10)
- ai (6)
- python (5)
- deep-learning (4)
- tensorflow (3)
- neural-networks (3)
- llm (3)
- writings (3)
- career (3)
- keras (2)
- collaborative-filtering (2)
- pytorch (2)
- anomaly-detection (1)
- gaussian-model (1)
- data-science (1)
- linear-regression (1)
- gradient-descent (1)
- logistic-regression (1)
- multiclass-classification (1)
- model-evaluation (1)
- overfitting (1)
- regularization (1)
- recommender-system (1)
- decision-tree (1)
- information-gain (1)
- reinforcement-learning (1)
- deep-q-learning (1)
- dqn (1)
- experience-replay (1)
- target-network (1)
- openai-gym (1)
- lunar-lander (1)
- langchain (1)
- langgraph (1)
- ai-agent (1)
- reflection (1)
- recommender-systems (1)
- movielens-dataset (1)
- coding-agent (1)
- context-window (1)
- reasoning-vs-coding (1)
- fine-tuning (1)
- lora (1)
- qlora (1)
- adapter (1)
- huggingface (1)
- github (1)
- guide (1)
- clustering (1)
- image-processing (1)
- k-means (1)
- unsupervised-learning (1)
- mysql (1)
- index (1)
- performance (1)
- database (1)
- psychology (1)
- logic (1)
- openclaw (1)
- telegram-bot (1)
- devops (1)
- self-hosted (1)
- transformer (1)
- attention (1)
- self-attention (1)
- bert (1)
- gpt (1)
- nlp (1)
- vibe-coding (1)
- java (1)
- Published on
深入解析 LangGraph 中的 Reflection Agents 技术，涵盖 Basic Reflection、Reflexion、LATS 三种核心方法，以及与 ReAct 的对比分析。包含完整代码示例和实战指南。- Published on
跟随本教程，学习如何应用深度Q学习（DQN）、经验回放和目标网络，从零开始训练一个能成功完成OpenAI Gym月球着陆器（Lunar Lander）任务的智能体。包含理论讲解、代码实现与结果展示。- Published on
在本实验中，您将实现协同过滤算法来构建一个电影推荐系统。通过理论讲解、代码实现和实践操作，您将深入理解协同过滤的核心原理。让我们开始吧！- Published on
本文完整实现了基于协同过滤的电影推荐系统，使用TensorFlow框架在MovieLens数据集上完成了从理论推导到工程实践的全流程。通过矩阵分解方法学习用户偏好和电影特征，实现了包含正则化的成本函数和梯度下降优化，最终生成个性化Top10推荐列表。实验显示模型能准确预测用户评分并推荐高口碑电影，为推荐系统入门提供了完整的实践范例。