- @hestendelin
简介
该用户还未填写简介
擅长的技术栈
可提供的服务
暂无可提供的服务
对于这个图像分类任务，我建议尝试 ResNet-50（预估准确率 92%），Vision Transformer（预估 93%），EfficientNet（预估 91%）。：即使与需要真训练的 AutoML 方法相比，AutoML-Agent 通过伪执行找到的方案性能也很接近（差异小于 5%），但速度快了 100 倍以上。（不同计划的数据专家、模型专家可以同时工作），而且“伪执行”本身很快，因此
Warm Up Cold-start Advertisements: Improving CTRPredictions via Learning to Learn ID EmbeddingsFeiyang Pan, Shuokai Li, Xiang Ao, Pingzhong Tang, Qing HeTsinghua University,Chinese Academy of Scienc..
Dual Graph enhanced Embedding Neural Network for CTRPredictionWei Guo, Rong Su, Renhao Tan, Huifeng Guo, Yingxue Zhang, Zhirong Liu,Ruiming Tang, Xiuqiang HeHuawei Noah’s Ark Labhttps://arxiv.org/pd..
Deep Learning for Click-Through Rate EstimationWeinan Zhang, Jiarui Qin, Wei Guo, Ruiming Tang...
DRN: A Deep Reinforcement Learning Framework for NewsRecommendationGuanjie Zheng, Fuzheng Zha...
Fi-GNN: Modeling Feature Interactions via Graph NeuralNetworks for CTR PredictionZekun Li, Z...
AutoInt: Automatic Feature Interaction Learning via Self-Attentive Neural NetworksWeiping Son...
Graph Neural Networks for Social RecommendationWenqi Fan,Yao Ma,Qing Li,Yuan He,Eric Zhao...
Finding Camouflaged Needle in a Haystack? PornographicProducts Detection via Berrypicking Tr...
[1]Deep Density DestructorsDavid I. InouyeCarnegie Mellon Universityhttp://proceedi...