Differentially private deep learning has recently witnessed advances in computational efficiency and privacy-utility trade-off. We explore whether further improvements along the two axes are possible and provide affirmative answers leveraging two instantiations of \emph{group-wise clipping}. To reduce the compute time overhead of private learning, we show that \emph{per-layer clipping}, where the gradient of each neural network layer is clipped separately, allows clipping to be performed in conjunction with backpropagation in differentially private optimization. This results in private learning that is as memory-efficient and almost as fast per training update as non-private learning for many workflows of interest. While per-layer clipping with constant thresholds tends to underperform standard flat clipping, per-layer clipping with adaptive thresholds matches or outperforms flat clipping under given training epoch constraints, hence attaining similar or better task performance within less wall time. To explore the limits of scaling (pretrained) models in differentially private deep learning, we privately fine-tune the 175 billion-parameter GPT-3. We bypass scaling challenges associated with clipping gradients that are distributed across multiple devices with \emph{per-device clipping} that clips the gradient of each model piece separately on its host device. Privately fine-tuning GPT-3 with per-device clipping achieves a task performance at $\epsilon=1$ better than what is attainable by non-privately fine-tuning the largest GPT-2 on a summarization task.
translated by 谷歌翻译
我们提出了一项合成任务,乐高(学习平等和小组操作),该任务封装了遵循推理链的问题,我们研究了变压器体系结构如何学习这项任务。我们特别注意数据效应,例如预处理(看似无关的NLP任务)和数据集组成(例如,训练和测试时间时的链长度不同),以及体系结构变体,例如重量绑定层或添加卷积组件。我们研究了受过训练的模型最终如何在任务中取得成功,尤其是我们能够在某种程度上(一定程度地)理解一些注意力头以及网络中的信息如何流动。基于这些观察结果,我们提出了一个假设,即在这里进行预训练仅是因为是智能初始化而不是网络中存储的深层知识。我们还观察到,在某些数据制度中,受过训练的变压器发现“快捷方式”解决方案遵循推理链,这阻碍了该模型将其推广到主要任务的简单变体的能力,而且我们发现人们可以防止适当的快捷方式架构修改或仔细的数据准备。在我们的发现的激励下,我们开始探索学习执行C程序的任务,在此过程中,对变压器进行了卷积修改,即在密钥/查询/值图中添加卷积结构,显示出令人鼓舞的优势。
translated by 谷歌翻译
我们为大规模训练的大规模训练语言模型提供了更简单,更稀疏,更快的算法,这些算法在许多标准的NLP任务上实现了最新的隐私与实用性权衡。我们为此问题提出了一个元框架,这是受高度参数效率方法进行微调成功的启发。我们的实验表明,这些方法的差异化适应能力在三个重要方面优于以前的私人算法:实用程序,隐私以及私人培训的计算和记忆成本。在许多经常研究的数据集中,私人模型的实用性接近了非私人模型的方法。例如,在MNLI数据集上,我们使用Roberta-large的准确度为87.8 \%$,使用Roberta-Base $ 83.5 \%$,其隐私预算为$ \ Epsilon = 6.7 $。相比之下,缺乏隐私限制,罗伯塔·莱格(Roberta-Large)的准确度为$ 90.2 \%$。我们的发现对于自然语言生成任务类似。与DART,GPT-2-SMALL,GPT-2中,GPT-2-MEDIUM,GPT-2-LARGE和GPT-2-XL的私人微调达到38.5、42.0、43.1和43.8($ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ 43.8) epsilon = 6.8,\ delta = $ 1E-5),而非私人基线为$ 48.1 $。我们所有的实验都表明,较大的模型更适合私人微调:虽然众所周知,它们旨在非优先实现卓越的准确性,但我们发现当引入隐私时,它们也更好地保持其准确性。
translated by 谷歌翻译
我们表明,具有内置关系偏差的深度学习模型可以带来利益来采样高效的学习,而无需依赖广泛的数据增强。所提出的单次分类模型以局部和成对注意的形式执行一对输入的关系匹配。我们的方法完美地解决了单次图像分类omniglot挑战。我们的模型超过人力学精度,以及以前的现有技术,没有数据增强。
translated by 谷歌翻译