SummaryDo you want to help build some of the largest and most consequential enterprise and customer technology systems in the world? Join Apple’s Information Systems and Technology (IS&T) organization. IS&T is the engine behind everything Apple
具身算法工程师(VLA+RL)-通用业务部 上海 校招 正式 技术 - 算法类 职位 ID:A121738 职位描述 VLA 端到端模型研发: 设计并训练端到端视觉-语言-动作模型(如 RT-2、OpenVLA、π0、Diffusion Policy、3D Diffusion Actor 等),实现从视觉输入与语言指令到机器人低维动作空间的直接映射。研究动作表示与生成机制: 动作 Token 化(Action Tokenization)、隐动作量化(Latent Action Quantization, LAPA)、扩散式动作生成(Diffusion Policy)、流匹配(Flow Matching)等前沿方案。探索 RL 与模仿学习(IL)的混合训练范式:利用人类演示数据初始化策略,再通过 RL 进行微调和鲁棒性增强。强化学习运控与策略优化: 包括 PPO、SAC、RLHF(Human Feedback for Robotics)、DAPG 等算法在关节级/任务级控制中的应用。模型部署与实时性优化:
SummaryDo you want to help build some of the largest and most consequential enterprise and customer technology systems in the world? Join Apple’s Information Systems and Technology (IS&T) organization. IS&T is the engine behind everything Apple