诺亚聚焦人工智能基础研究，秉承“一杯咖啡吸收宇宙能量”的理念，与业界开展广泛的交流合作，积极参与开源，通过开源项目回馈社区，共同推进人工智能基础研究的持续突破。
开源项目如下：
计算机视觉
- 【Full-Stack-Filters】Pytorch code for paper: Full-Stack Filters to Build Minimum Viable CNNs
- 【Versatile-Filters】Pytorch codefor paper: Learning Versatile Filters for Efficient Convolutional NeuralNetworks (NeurIPS 2018)
- 【DAFL】A Pytorch implementation of "Data-Free Learning of Student Networks" (ICCV2019)
- 【GAN-pruning】A Pytorch implementation of "Co-Evolutionary Compression for Unpaired ImageTranslation"(ICCV 2019)
- 【LegoNet】A Pytorch implementation of "LegoNet: Efficient Convolutional Neural Networks with Lego Filters"(ICML 2019)
决策推理
- 【StreamDM】Stream Data Mining Library for Spark Streaming
- 【StreamDM-Cpp】Stream Machine Learning in C++
- 【BGCN】A Tensorflow implementation of "Bayesian Graph Convolutional Neural Networks" (AAAI 2019).
- 【BHT-ARIMA】A Python implementation of “Block Hankel Tensor ARIMA for Multiple Short Time Series Forecasting”(AAAI-20)
自然语言处理
- 【TinyBERT】TinyBERT is a compressed BERT model which achieves 7.5x smaller and 9.4x faster on inference.
- 【NEZHA】NEZHA is a pretrained Chinese language model which achieves the state-of-the-art performances on several Chinese NLP tasks
更多详细的信息欢迎访问诺亚方舟实验室官方网站查询。
推荐阅读
文章首发知乎，更多深度模型压缩相关的文章请关注深度学习压缩模型论文专栏。
