14 Matching Annotations
  1. Dec 2023
    1. 先在 Evernote 中選擇需要的筆記、記事本,匯出成 enex 檔案。

      Confirming Upnote support ENEX import from Evernote. I'm ready to pay $1/month for this!

  2. Mar 2021
  3. May 2020
    1. AWS Batch enables developers, scientists, and engineers to easily and efficiently run hundreds of thousands of batch computing jobs on AWS. AWS Batch dynamically provisions the optimal quantity and type of compute resources (e.g., CPU or memory optimized instances) based on the volume and specific resource requirements of the batch jobs submitted.
  4. Mar 2019
  5. Feb 2019
    1. Interplay Between Optimization and Generalization of Stochastic Gradient Descent with Covariance Noise

      一个有趣的事实:batch-size 对训练收敛和模型泛化表现是有影响的,batch-size 越大,收敛越好,泛化变差。。。

    2. LocalNorm: Robust Image Classification through Dynamically Regularized Normalization

      提出了新的 LocalNorm。既然Norm都玩得这么嗨了,看来接下来就可以研究小 GeneralizedNorm 或者 AnyRandomNorm 啥的了。。。[doge]

    3. Fixup Initialization: Residual Learning Without Normalization

      关于拟合的表现,Regularization 和 BN 的设计总是很微妙,尤其是 learning rate 再掺和进来以后。此 paper 的作者也就相关问题结合自己的文章在 Reddit 上有所讨论。

  6. Dec 2018
    1. Generalized Batch Normalization: Towards Accelerating Deep Neural Networks

      核心是这么一句话: Generalized Batch Normalization (GBN) to be identical to conventional BN but with

      1. standard deviation replaced by a more general deviation measure D(x)

      2. and the mean replaced by a corresponding statistic S(x).

    2. How Does Batch Normalization Help Optimization? (No, It Is Not About Internal Covariate Shift)


  7. Nov 2018
    1. Revisiting Small Batch Training for Deep Neural Networks

      这篇文章简而言之就是mini-batch sizes取得尽可能小一些可能比较好。自己瞅了一眼正在写的 paper,这不禁让我小肝微微一颤,心想:还是下次再把 batch-size 取得小一点吧。。。[挖鼻] ​​​​

    2. Don't Use Large Mini-Batches, Use Local SGD

      最近(2018/8)在听数学与系统科学的非凸最优化进展时候,李博士就讲过:现在其实不太欣赏变 learning rate 了,反而逐步从 SGD 到 MGD 再到 GD 的方式,提高 batch-size 会有更好的优化效果!

  8. Oct 2018
    1. Approximate Fisher Information Matrix to Characterise the Training of Deep Neural Networks

      深度神经网络训练(收敛/泛化性能)的近似Fisher信息矩阵表征,可自动优化mini-batch size/learning rate

      挺有趣的 paper,提出了从 Fisher 矩阵抽象出新的量用来衡量训练过程中的模型表现,来优化mini-batch sizes and learning rates | 另外 paper 中的figure画的很好看 | 作者认为逐步增加batch sizes的传统理解只是partially true,存在逐步递减该 size 来提高 model 收敛和泛化能力的可能。

  9. Feb 2018
  10. Mar 2016