Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/07/12/tpc.18.00785.DC2/tpc18.00785.PeerReviewReport.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/04/09/tpc.18.00778.DC2/tpc18.00778.PeerReview.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/04/12/tpc.18.00946.DC2/tpc18.00946.PeerReview.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/04/17/tpc.18.00938.DC2/tpc18.00938.PeerReview.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/07/13/tpc.18.00606.DC2/tpc18.00606.PRR-Griffiths.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/07/02/tpc.19.00132.DC2/tpc19.00132.PeerReviewReport.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/07/13/tpc.19.00033.DC2/tpc19.00033.PRR-Niittyla.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/05/18/tpc.18.00043.DC2/tpc18.00043.PeerReviewReport.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/05/07/tpc.19.00047.DC2/tpc19.00047.PeerReview.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/05/29/tpc.19.00235.DC2/tpc19.00235.PeerReviewReport.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/07/13/tpc.18.00918.DC2/tpc18.00918.PRR-Yin.pdf

Check out the peer review report for this article: http://www.plantcell.org/content/plantcell/suppl/2019/05/30/tpc.19.00015.DC2/tpc19.00015.PeerReview.pdf

Check out the peer review report here http://www.plantcell.org/content/plantcell/suppl/2019/07/13/tpc.19.00069.DC2/tpc19.00069.PRR-Genschik.pdf

Check out the peer review report here http://www.plantcell.org/content/plantcell/suppl/2019/06/12/tpc.18.00952.DC2/tpc18.00952.PeerReview.pdf

Check out the peer review report here http://www.plantcell.org/content/plantcell/suppl/2019/07/02/tpc.18.00840.DC2/tpc18.00840.PeerReviewReport.pdf

Check out the peer review report here http://www.plantcell.org/content/plantcell/suppl/2019/07/02/tpc.18.00706.DC2/tpc18.00706.PeerReviewReport.pdf

Check out the peer review report here https://doi.org/10.1105/tpc.19.00001

Check out the peer review report here http://www.plantcell.org/content/plantcell/suppl/2019/07/03/tpc.19.00081.DC2/tpc19.00081.PeerReviewReport.pdf

My post-publication peer review reports a re-analysis of this article's data, adds additional figures, and questions whether what the authors. I ask the authors to more fully release and analyse their coded data.

My post-publication peer review and re-analysis identifies problems with the included temperature range and the linearity of effects.

connection: we shouldn't seek external approval because that approval is unreliable.

a corollary is that by doing so, we get more external approval. We are more grounded.

【导读】变分自编码器（VAE）是重要的生成式模型。与生成式对抗网络（GAN）类似，VAE也可以被用来生成逼真的图像和文本信息，但VAE的思想却与GAN有很大的区别。本文介绍Arxiv上的一篇93页VAE导论，该导论包含大量的公式推导和图示。

近几年来，生成式对抗网络（GAN）吸引了大量科研人员和工程师的关注。然而除了GAN，变分自编码器（VAE）也是这几年较为火热的重要的生成式模型。与GAN的利用生成器和判别器进行对抗的思路不同，VAE的核心组件是自编码器和KL散度约束。

异常检测的综述

Evaluation of technology enhanced learning programs for health care professionals: systematic review This article is included because it is a systematic review. It is presented in academic language. The intention is to evaluate the quality of the articles themselves, not to guide e-learning development. Criteria for evaluating articles was established in advance. The utility of the article for my purposes may be a new search term, continuous professional development. rating 2/5

New Media Consortium Horizon Report This page provides a link to the annual Horizon Report. The report becomes available late in the year. The report identifies emerging technologies that are likely to be influential and describes the timeline and prospective impact for each. Unlike the link to top learning tools that anyone can use, the technologies listed here may be beyond the ability of the average trainer to implement. While it is informative and perhaps a good idea to stay abreast of these listings, it is not necessarily something that the average instructional designer can apply. Rating: 3/5

深度强化学习综述

深度文本匹配综述

区块链技术：架构及进展

A Sensitivity Analysis of (and Practitioners’ Guide to) Convolutional Neural Networks for Sentence Classification

A Comparative Study of Neural Network Models for Sentence Classification

In his commentary, Alex Holcombe makes the argument that only ‘one or two exemplars of a color category’ are typically examined in color studies, and this is problematic because a color such as ‘red’ is a category, not a single hue.

Although in some fields it is very important to examine a range of stimuli, and in general examining the generalizability of findings has an important place in research lines, I do not think that currently this issue is a pressing concern in color psychology. Small variations in hue and brightness naturally occur in online studies, and these are assumed not to matter for the underlying mechanism. Schietecat, Lakens, IJsselsteijn, and De Kort (2018) write: “In addition, we conducted Experiments 1 and 3 in a laboratory environment, but Experiments 2, 4, and 5 were conducted in participants’ homes with an internet-based method. Therefore, we could not be completely sure that the presentation of the stimuli on their personal computers was identical for every participant in those experiments. However, we expected that the impact of these variations on our results is not substantial. The labels of the IAT (i.e., red vs blue) increased the salience of the relevant hue dimension, and we do not expect our results to hold for very specific hues, but for colors that are broadly categorized as red, blue, and green. The similar associative patterns across Experiments 2 and 3 seem to support this expectation.”

We wrote this because there is nothing specific about the hue that is expected to drive the effects in association based accounts of psychological effects of colors. If the color ‘red’ is associated with specific concepts (and the work by Schietecat at all supports the idea that red can activate associations related to either activity and evaluation, such as aggression or enthusiasm, depending on the context). This means that the crucial role of the stimulus is to activate the association with ‘red’, no the perceptual stimulation of the eye in any specific way. The critical manipulation check would thus be is people categorize a stimulus as ‘red’. As long as this is satisfied, we can assume the concept ‘red’ is activated, which can then activate related associations, depending on the context.

Obviously, the author is correct that there are benefits in testing multiple variations of the color ‘red’ to demonstrate the generalizability of observed effects. However, the authors is writing too much as a perception researcher I fear. If there is a strong theoretical reason to assume slightly different hues and chromas will not matter (because as long as a color is recognized as ‘red’ it will activate specific associations) the research priority of varying colors is much lower than in other fields (e.g., research on human faces) where it is more plausible that the specifics of the stimuli matter. A similar argument holds for the question whether “any link is specifically to red, rather than extending to green, yellow, purple, and brown”. This is too a-theoretical, and even though not all color research has been replicable, and many studies suffered from problems identified during the replication crisis, the theoretical models are still plausible, and specific to predictions about certain hues. We know quite a lot about color associations for prototypical colors in terms of their associations with valence and activity (e.g., Russell & Mehrabian, 1977) and this can be used to make more specific predictions than to a-theoretically test the entire color spectrum.

Indeed, across the literature many slightly different variations of red are used, or in online studies (Schietecat et al., 2018) studies have been performed online, where different computer screens will naturally lead to some variation in the exact colors presented. This doesn’t mean that more dedicated exploration of the boundaries of these effects can be worthwhile in the future. But currently, the literature is more focused on examining whether these effects are reliable to begin with, and explaining basic questions about their context dependency, than that they are concerned about testing the range of hues for which effects can be observed. So, although in principle it is often true that the generalizability of effects is understudies and deserved more attention, it is not color psychology’s most pressing concern, because we have theoretical predictions about specific colors, and because theoretically as long as a color activates the concept (e.g., ‘red’), the associated concepts that influence subsequent psychological responses are assumed to be activated, irrespective of minor differences in for example hue or brightness.

Daniel Lakens

References

Russell, J. A., & Mehrabian, A. (1977). Evidence for a three-factor theory of emotions. Journal of Research in Personality, 11(3), 273–294. DOI: https://doi.org/10.1016/0092-6566(77)90037-X Schietecat, A. C., Lakens, D., IJsselsteijn, W. A., & Kort, Y. A. W. de. (2018). Predicting Context-dependent Cross-modal Associations with Dimension-specific Polarity Attributions. Part 2: Red and Valence. Collabra: Psychology, 4(1). https://doi.org/10.1525/COLLABRA.126

Question Answering, Task-Oriented Dialogues and Social Chatbots

The present paper surveys neural approaches to conversational AI that have beendeveloped in the last few years. We group conversational systems into three cat-egories: (1) question answering agents, (2) task-oriented dialogue agents, and(3) chatbots. For each category, we present a review of state-of-the-art neuralapproaches, draw the connection between them and traditional approaches, anddiscuss the progress that has been made and challenges still being faced, usingspecific systems and models as case studies

任务型人机对话系统中的认知技术

非常简单的介绍，可以看作是一个文献阅读索引

detail :

• http://opendialogue.miulab.tw

• https://www.csie.ntu.edu.tw/~yvchen/doc/OpenDialogue_Tutorial_ISCSLP18.pdf

• https://www.csie.ntu.edu.tw/~yvchen/doc/OpenDialogue_Tutorial_IJCNLP.pdf

Open-Domain Neural Dialogue Systems

Dialog System & Technology Challenge 6 Overview of Track 1 - End-to-End Goal-Oriented Dialog learning

End-to-end dialog learning is an important research subject inthe domain of conversational systems. The primary task consistsin learning a dialog policy from transactional dialogs of a givendomain. In this context, usable datasets are needed to evaluatelearning approaches, yet remain scarce. For this challenge, atransaction dialog dataset has been produced using a dialogsimulation framework developed and released by Facebook AIResearch. Overall, nine teams participated in the challenge. Inthis report, we describe the task and the dataset. Then, we specifythe evaluation metrics for the challenge. Finally, the results ofthe submitted runs of the participants are detailed.

对话系统任务综述与基于POMDP的对话系统

Survey on Chatbot Design Techniques in Speech Conversation Systems 2015、 没啥意义

we define conversation-turns per session (CPS) as the success metric for social chatbots. The larger the CPS is, the better engaged the social chatbot is.

【导读】图像超分（SR, Super-Resolution）图像处理及数据非常重要的应用方向，主要目标在于增强原始图像与视频的分辨率精度。最近这些年，非常多的图像超分问题研究均采用了深度学习的架构，本篇综述希望通过全面回顾基于深度学习的图像超分方法，帮助大家快速了解这一领域的最新动态。

介绍：

图像超分问题，主要目标是图像处理技术中的重要研究方向，主要目标是将图片从低分辨率恢复到高分辨率的图像。这类方法具有非常广泛的应用价值，比如医疗影像、安防等等。通常来说，这一问题非常具有挑战性，因为总是存在多种对应于相同低分辨率图像的高分辨率映射。在以往的研究中，提出了一些传统的超分方法，包括基于预测的方法，基于边的方法，以及稀疏表示方法等等。

随着深度学习技术的快速发展，基于深度学习的图像超分方法已经被开发出来，并且在多个测试任务上，取得了目前最优的性能效果。多种深度学习方法已经被应用到具体的图像超分任务中，从早期的卷积神经网络（SRCNN）到最近提出的基于GAN的超分方法。通常来说，使用了深度学习的超分算法类中，存在着以下几种不同之处：网络架构不同、损失函数不同、学习原理、策略不同等。

在这篇文章中，主要给出了在超分算法中使用深度学习技术的一些优势。虽然存在着一些其他关于超分算法的综述，但我们不同之处在于主要关注基于深度学习的超分算法，而不像其他早期工作那样关注于传统的超分算法综述。本篇综述给出了一个统一的深度学习视角，来回顾最近的超分技术进展。

本文的主要贡献包括三个方面:

1. 给出了一个综合的基于深度学习的图像超分技术综述，包括问题设置、数据集、性能度量、一组基于深度学习的图像超分方法集合，特定领域的图像超分方法应用等等。

2. 为最近基于深度学习的图像超分算法提供了系统性、结构化的视角，并总结了高效图像超分解决方案中的优势于劣势。

3. 我们讨论了这个领域的挑战与开放问题，并总结了最近的新趋势与未来的发展方向。

无监督的视觉表示学习在计算机视觉研究中仍然是一个很大程度上未解决的问题。在最近提出的用于无监督学习视觉表示的方法中，一类自我监督技术在许多具有挑战性的基准上实现了卓越的性能。已经研究了大量的自我监督学习的前提任务，但其他重要的方面，如卷积神经网络（CNN）的选择，并没有得到同等的关注。因此，我们重新审视了许多以前提出的自我监督模型，进行彻底的大规模研究，结果发现了多个关键的问题。我们挑战了自我监督的视觉表现学习中的一些常见实践，并观察到CNN设计的标准配方并不总是转化为自我监督的表征学习。作为我们研究的一部分，我们大大提高了先前提出的技术的性能，并且大大优于以前发布的最先进的结果。

GAN 自 2014 年诞生至今也有 4 个多年头了，大量围绕 GAN 展开的文章被发表在各大期刊和会议，以改进和分析 GAN 的数学研究、提高 GAN 的生成质量研究、GAN 在图像生成上的应用（指定图像合成、文本到图像，图像到图像、视频）以及 GAN 在 NLP 和其它领域的应用。图像生成是研究最多的，并且该领域的研究已经证明了在图像合成中使用 GAN 的巨大潜力。本文对 GAN 在图像生成应用做个综述。

For a while I've been doing PESOS from reading.am to my website privately. Then a day or so later I come back to the piece to think about it again and post any additional thoughts, add tags, etc. I often find that things I missed the first time around manage to resurface. Unless I've got a good reason not to I usually then publish it.

Biomed Central and ResearchSquare have partnered on a project called InReview, which enables community feedback in parallel with traditional peer review. More on that project here.

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

深度卷积神经网络（CNN）是一种特殊类型的神经网络，在各种竞赛基准上表现出了当前最优结果。深度 CNN 架构在挑战性基准任务比赛中实现的高性能表明，创新的架构理念以及参数优化可以提高 CNN 在各种视觉相关任务上的性能。本综述将最近的 CNN 架构创新分为七个不同的类别，分别基于空间利用、深度、多路径、宽度、特征图利用、通道提升和注意力。

感觉此文于我而言真正有价值的恐怕只有文末附录的 Dataset tables 汇总整理了。。。。。

And the combination of community feedback on preprints with traditional and post-publication peer review through collaborative annotation is catching on with a variety of publishers. See InReview by BMC and ResearchSquare. Also COS preprint servers such as SocArXiv and Psyarxiv.

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This paper has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

This article has a Peer Review Report

Apakah hasil peninjauan ini terbuka untuk umum dan diberikan juga kepada peneliti? Maaf kalau saya keliru, saya setiap tahun mengirimkan proposal ke Kemristekdikti, tapi hasil peninjauan secara lengkap belum pernah saya terima.

Please find a Peer Review Report here.

The report shows the major requests for revision and author responses. Minor comments for revision and miscellaneous correspondence are not included. The original format may not be reflected in this compilation, but the reviewer comments and author responses are not edited, except to correct minor typographical or spelling errors that could be a source of ambiguity.

果断收藏并且要细读下。。。Paper Summary 准备！

这可是对 CNN 可视化方法的 review 啊！

一篇很棒的综述，专门说 CNN 的可视化的！要好好读读了！

Paper Summary 准备！

一篇印度人写的对抗性防御的综述paper。

看到 Goodfellow 亲自转推了此文~~

很重要的 review，准备个 Paper Summary

一篇谈深度学习模型泛化能力的 review paper~ 不错。。。就是很理论。。很泛泛。。。

这个文与我的课题貌似相当相关！

准备好好写一个 Paper Summary 为好~

• 这是一篇很牛的 review。
• 被引用数已经3000+。
• 总共有88页，但正文仅35页。(88 pages, 888 references)
• 详述了深度学习在神经网络历史与发展，很值得参考！

一篇关于无标签正样本学习(PU Learning)的综述。自己应该好好读读，在现实情况下，这种类别很多很重要。。。

此文综述了最近 Random Neural Network (RNN) 模型的应用，文末的 summary 和 future work，可以看到 RNN 已经在很多方面有着不错的表现，但是还有更多的问题需要进一步探讨和研究。

这个 review 在一定程度上指出了当前前沿 ML 研究的趋势和局限性，重要的是它为未来应当着力发展和探究的课题给出了几个方向。

如题。

review 一篇。。。。

这是一篇很有 review 气质的 paper，对GAN 和对抗方法等做了介绍（在生物医药领域中），也谈论了这些技术应用的优势和劣势。

对我来说，这是一篇很适合快速入门GAN应用的 paper。

一篇关于模型选择的综述文章。涉及信号处理，图像处理等等多方面数据信息的处理。发表在信号处理的期刊杂志上。

文中关于模型选择的大概念方向，和数学表示，是值得好好阅读的。

不仅综述了迁移学习的现状，也对其进行了分类。同时，还给出了“深度迁移”的概念，强调了待迁移的两个学习任务之间的非线性关系。其实这也很自然，我们本来对线性的“相似”学习任务迁移就兴趣一般，也没多大研究意义。。。。

Deepmind 的这篇文有点水，只能归为 review 了。涉及了很多数据分析的方法和 DNN 的概念，当然也未跑题的谈论了生物/人工神经网络之间的爱恨纠葛。

Synergy 这个词特别有趣，译为：协同增效作用。

I can relate this to my earlier experiences. Active involvement and students' engagement is guaranteed if they are tasked with solving actual physics problems related to the materials being taught before.

这个综述的角度比较有趣，既易读也很有启发性，将模型分为三大类，它们之间的关系总结为下图：

当然并不是所有模型都被囊括在其中，比如玻尔兹曼机等~

【Review】【文章摘要】

随着深度学习逐渐成为机器学习领域的引领技术，机器学习领域正在见证其黄金时代。深度学习使用多个层来表示数据的抽象，以构建计算模型。一些关键的使能深度学习算法，如生成对抗网络，卷积神经网络和模型迁移，彻底改变了我们对信息处理的方法。然而，在这个极其快节奏的领域背后存在着一种理解，因为它以前从未以多视角展示过。缺乏核心理解使得这些强大的方法成为黑盒机器，从根本上抑制了深度学习发展。此外，深度学习一再被视为机器学习中所有绊脚石的银弹，这远非事实。本文全面回顾了视觉，音频和文本处理方面的历史和最新技术方法;社交网络分析;和自然语言处理，然后深入分析深度学习应用中的突破性进展。还开展了审查深度学习所面临的问题，如无监督学习，黑盒模型和在线学习，并说明如何将这些挑战转化为多产的未来研究途径。

1) read the paper thoroughly 2) summarize your main points 3) relevant past work 4) significance of contribution and benefit 5) coverage of all the criteria 6) review "as is" 7) polite, temperate language

good meta-review -> weight more heavily from reviews, and why

good meta-review: -> consider all of the reviewers' opinions

good review: 1) make great suggestions for how the authors could improve the articulation or organization of their work

good review: 1) raise smart and tough questions which the authors can address in their revisions 2) raise fresh considerations or new aspects of a design space that the authors hadn't fully fleshed out

argument + results + demonstration + related word + the field, all of them should tight together!

good review: 1) raise whole new perspectives and angles of contributions 2) propose connections to literature that the author might not have been aware of

Good Review 1) reflect on the contributions of the work 2) discuss the weaknesses and limitations in a positive manner 3) clearly call out the strengths and utility of the work

The Eden Club - Exclusive Golf Club Membership | International Private Members Club

The Eden Club is an international private members' club providing three very special dimensions: the most luxurious private members club in St Andrews, Scotland – the home of golf; an outstanding schedule of annual events and a unique Secretariat service.

This is before the PAW method came along (Blochl '94), so probably nothing method-specific.

open peer review model

Suggests that RBPmap is a good choice, provided one uses the weighted rank score accounting for clustering and the conservation filter. mCarts by Zhang et al. applied to the ENCODE CLIP data would complement RBPmap, but what would be even better is if ENCODE had already applied it to their data and can provide the binding site predictions.

Discerning the available values of epub:type is a tangle... The referenced epub:type related section in the ContentDocs32 document references the Vocabulary Association paragraph which in turn references the Vocabulary Association Mechanisms section within this current document. From that section the default vocabulary can/may be found, but there is no reference there (or in any of these other sections) to the available epub:type values nor to their "meaning"/value to the author or implementer.

However, just below this in the example (and above in relation to pagebreak there is a link directly to the defining defaults in the EPUB-SSV.

It would be helpful for authors and implementers to know first about the SSV list, and then about the ability to extend the vocabulary.

If this SHOULD is not heeded...what happens? How is the page list calculated if there is a hierarchy present?

This makes it sound like there is an HTML element with a tag name of <toc nav> rather than <nav epub:type="toc"> (which seems to be what's intended).

The landmarks example farther down is clearer--though the wording there of "the landmark nav element" is equally confusing.

There remains only a nav element, but of varying types.

This line is confusing because the actor (author or reading system) is unclear. The following two lines are much clearer on what actor is being restricted by this specification.

This line feels specific to the Reading System requirements. Consequently, perhaps it's best to integrate it into the sentence just following.

Link resolve to any defined algorithms for resolving resources within an EPUB.

This requirement is clearer than the one just previous and also seems to obviate the earlier one.

However, this one is confusing limiting the textual expression to just the title attribute, and also being vague of which element is associated to the MUST--is it this surrounding a or span? or the contained "embedded content" element?

Implied requirement? When is this necessary and what does the concatenation and normalization?

Where is this "visual display" taking place? Is this when the document is viewed as a spine element? or is this when the content is extracted and presented by the Reading System?

Rephrase "while a child span element" and perhaps move to it's own sentence parallel to the "A child a element"

Reference the li element here for consistency.

Link to this, if defined.

Is "non-browser context" defined? or its inverse?

It's not clear if "browser" here is equivalent to "Reading System" or to a Web browser or something else.

Used by? It's not clear who the actor is here, the author/developer of the EPUB or the Reading System.

Is the "table of contents" a specific type of custom control? If so, a link would help here.

But it does have effect on "spine rendering"? Perhaps add a link for spine rendering.

Is "data island" a specific, defined thing from a previous EPUB spec or another adjacent spec?

Link to a definition of "custom control" if one exists.

Not sure what a "typical XML data island" is either...so some links or even rewording (or removing?) might help others unfamiliar with the phrase.

Interesting connection here between the efficiency mandate and the "talent means no effort required" attitude we see in many school settings.

I don't think we're pretending. I know I'm not!

Do your colleagues who read your work, annotate it, and comment on it not count as peer-review?

Am I wasting my time by annotating all of this? :) (I don't think so...)

May 31: Lecture 6 / Screencast Video Link

Literature Review

I have reviewed many papers and I summarized those papers in my paper in the form of a table which is already an existing theory . So how can I defend that my paper is different from other papers?

Wow. This could have a chilling effect on reviews for certain subjects.

Keeping Up with...Open Peer Review

redact, change verbiage

Este párrafo muestra el "ingrediente" de los 8 ingredientes de la pedagogia abierta de Peer review al hacer que los alumnos colaboren en el proceso de evaluación.