小米集团
校招布拉格直辖市 · 校招 · 2027 届 · 本科
长期有效
Responsibilities Overview: We welcome fresh graduates who are interested in the retail industry to join us and create a better future together! Multi-channel retail management: Learn and master the company's retail channel operation model, participate in channel analysis and strategy adjustment. New retail model exploration: In-depth understanding of online and offline integration and new retail trends, participate in new retail project practice. Strategic operation analysis: Use data analytics to mine market opportunities and risks, and provide data support and strategy suggestions for retail business. Future opportunities: 1. Deeply participate in the formulation and execution of multi-channel retail strategies. 2. Participate in the innovative practice of international new retail and the development of the company's retail business. 3. Fully participate in all aspects of retail management to ensure the stable development of the business。
上传简历,AI 会对照这份 JD 指出你的差距和需要补强的地方
基于 算法 岗位的常见面试问题整理
请解释 Transformer 模型的自注意力机制(Self-Attention)
1. Self-Attention = 序列中每个 token 与所有 token 计算相关性权重 2. Q=Wq·x, K=Wk·x, V=Wv·x; Attention(Q,K,V)=softmax(QK^T/√d_k)V 3. 除以 √d_k 防止点积过大导致 softmax 梯度消失 4. 多头注意力:多组 QKV 并行,拼接后线性变换,捕获不同子空间特征 5. 对比 RNN:可并行计算,捕获长距离依赖
你做过的最有深度的算法/模型项目是什么
1. 一句话说清任务:分类/生成/检索/推荐 2. 模型选型:为什么选这个而非那个 3. 数据:数据量、来源、清洗策略 4. 创新点:你做的改进是什么(不是调参) 5. 结果:核心指标提升(准确率/召回/延迟)
过拟合是什么?怎么判断和解决
1. 过拟合=训练集表现好、验证集表现差,模型记住了噪声 2. 判断:train loss 持续下降而 val loss 上升 3. 解决:更多数据 / 正则化(L1/L2) / Dropout / 早停 / 数据增强 / 降低模型复杂度
解释梯度下降的原理,SGD 和 Adam 的区别
1. 梯度下降:沿负梯度方向更新参数,使损失降低 2. SGD:每次用随机小批量估算梯度,引入随机性帮助跳出局部最优 3. Adam = Momentum + RMSProp:自适应学习率 + 动量加速 4. Adam 收敛快但可能泛化不如 SGD;CV 常用 SGD,NLP 多用 Adam
给你 10 万条未标注的文本数据,如何找出其中 10 个主题
1. 预处理:分词、去停用词 2. 向量化:TF-IDF 或 Sentence Embedding 3. 聚类:K-Means(需指定 K=10)或 HDBSCAN(自动确定类数) 4. 或 LDA 主题模型直接输出主题-词分布 5. 评估:看每个簇的高频词判断主题合理性,迭代调整
更多面试题库功能即将上线
信息来源:企业官方招聘页。投递前请以官网信息为准。