Self-supervised learning (SSL) has emerged as a promising paradigm that
presents self-generated supervisory signals to real-world problems, bypassing
the extensive manual labeling burden. SSL is especially attractive for
unsupervised tasks such as anomaly detection, where labeled anomalies are often
nonexistent and costly to obtain. While self-supervised anomaly detection
(SSAD) has seen a recent surge of interest, the literature has failed to treat
data augmentation as a hyperparameter. Meanwhile, recent works have reported
that the choice of augmentation has significant impact on detection
performance. In this paper, we introduce ST-SSAD (Self-Tuning Self-Supervised
Anomaly Detection), the first systematic approach to SSAD in regards to
rigorously tuning augmentation. To this end, our work presents two key
contributions. The first is a new unsupervised validation loss that quantifies
the alignment between the augmented training data and the (unlabeled) test
data. In principle we adopt transduction, quantifying the extent to which
augmentation mimics the true anomaly-generating mechanism, in contrast to
augmenting data with arbitrary pseudo anomalies without regard to test data.
Second, we present new differentiable augmentation functions, allowing data
augmentation hyperparameter(s) to be tuned end-to-end via our proposed
validation loss. Experiments on two testbeds with semantic class anomalies and
subtle industrial defects show that systematically tuning augmentation offers
significant performance gains over current practices.

本文介绍了基于自我调整自监督异常检测（ST-SSAD）的新方法，该方法通过数据扩充来调整超参数，从而在语义类异常和微妙的工业缺陷中实现了显著的性能提升。

自监督异常检测的端到端数据增强超参数调整

End-to-End Augmentation Hyperparameter Tuning for Self-Supervised  Anomaly Detection

We propose DocFormerv2, a multi-modal transformer for Visual Document
Understanding (VDU). The VDU domain entails understanding documents (beyond
mere OCR predictions) e.g., extracting information from a form, VQA for
documents and other tasks. VDU is challenging as it needs a model to make sense
of multiple modalities (visual, language and spatial) to make a prediction. Our
approach, termed DocFormerv2 is an encoder-decoder transformer which takes as
input - vision, language and spatial features. DocFormerv2 is pre-trained with
unsupervised tasks employed asymmetrically i.e., two novel document tasks on
encoder and one on the auto-regressive decoder. The unsupervised tasks have
been carefully designed to ensure that the pre-training encourages
local-feature alignment between multiple modalities. DocFormerv2 when evaluated
on nine datasets shows state-of-the-art performance over strong baselines e.g.
TabFact (4.3%), InfoVQA (1.4%), FUNSD (1%). Furthermore, to show generalization
capabilities, on three VQA tasks involving scene-text, Doc- Formerv2
outperforms previous comparably-sized models and even does better than much
larger models (such as GIT2, PaLi and Flamingo) on some tasks. Extensive
ablations show that due to its pre-training, DocFormerv2 understands multiple
modalities better than prior-art in VDU.

提出了一种多模态 Transformer 模型 - DocFormerv2，应用于视觉文档理解领域，在多个数据集上都取得了最先进的结果，并通过大量消融实验证明了其预训练和无监督任务设计的有效性。

DocFormerv2: 文档理解的本地特征

DocFormerv2: Local Features for Document Understanding

Astronomical surveys of celestial sources produce streams of noisy time
series measuring flux versus time ("light curves"). Unlike in many other
physical domains, however, large (and source-specific) temporal gaps in data
arise naturally due to intranight cadence choices as well as diurnal and
seasonal constraints. With nightly observations of millions of variable stars
and transients from upcoming surveys, efficient and accurate discovery and
classification techniques on noisy, irregularly sampled data must be employed
with minimal human-in-the-loop involvement. Machine learning for inference
tasks on such data traditionally requires the laborious hand-coding of
domain-specific numerical summaries of raw data ("features"). Here we present a
novel unsupervised autoencoding recurrent neural network (RNN) that makes
explicit use of sampling times and known heteroskedastic noise properties. When
trained on optical variable star catalogs, this network produces supervised
classification models that rival other best-in-class approaches. We find that
autoencoded features learned on one time-domain survey perform nearly as well
when applied to another survey. These networks can continue to learn from new
unlabeled observations and may be used in other unsupervised tasks such as
forecasting and anomaly detection.

本文介绍了一种基于自编码递归神经网络的自监督分类模型，它通过对光学变星数据的学习提取特征，能够在天文调查中有效地发现和分类变量星和瞬变现象，并可应用于其他无监督任务，例如预测和异常检测。

用于不均匀采样变星分类的循环神经网络

A recurrent neural network for classification of unevenly sampled  variable stars

Bidirectional recurrent neural networks (RNN) are trained to predict both in
the positive and negative time directions simultaneously. They have not been
used commonly in unsupervised tasks, because a probabilistic interpretation of
the model has been difficult. Recently, two different frameworks, GSN and NADE,
provide a connection between reconstruction and probabilistic modeling, which
makes the interpretation possible. As far as we know, neither GSN or NADE have
been studied in the context of time series before. As an example of an
unsupervised task, we study the problem of filling in gaps in high-dimensional
time series with complex dynamics. Although unidirectional RNNs have recently
been trained successfully to model such time series, inference in the negative
time direction is non-trivial. We propose two probabilistic interpretations of
bidirectional RNNs that can be used to reconstruct missing gaps efficiently.
Our experiments on text data show that both proposed methods are much more
accurate than unidirectional reconstructions, although a bit less accurate than
a computationally complex bidirectional Bayesian inference on the
unidirectional RNN. We also provide results on music data for which the
Bayesian inference is computationally infeasible, demonstrating the scalability
of the proposed methods.

本文提出两种概率解释双向循环神经网络并应用于高维时间序列数据中的缺失值填充，在文本数据表现比单向推断更加准确，而在音乐数据中则具有可扩展性。