投稿前的28个细节检查清单

Have you --------------?

1 spell checked the whole paper?

2 rewritten your paper at least three times?

3 stopped working on your paper for a day?

To see and hear your words better when you look at it again.

4 put your main point into your title?

5 put your most significant findings first? Your gold first?

6 made it easy for readers to tell from your title and your abstract what is new and important about your work?

7 made sure all your verbs and subjects match?

Plural verbs have plural subjects? Singular verbs have singular subjects?

8 spelled the same name exactly the same way throughout your paper?

NOT “Bragg. . . .Brag”

BUT BRAGG every time!

9 asked a friend or a native English speaker to read your paper and tell you what’s confusing or unclear?

10 changed nouns to adjectives when necessary?

NOT “three dimension ordered array”

BUT “three dimensional ordered array”

11 used active verbs rather than passive verbs as much as possible?

NOT “The effects were investigated. . .”

BUT “We investigated the effects”

12 found a way to break a sentence that’s much too long (over 4 lines) into two sentences?

13 broken up a paragraph that is way too long into two paragraphs?

14 tried to reduce your paper to its skeleton by making a one page outline of it?

Might help you see if your paper is organized in the best possible way.

 

15 Made all equivalent items look the same?

NOT “scheme 1 . . . .Scheme 2”

BUT “Scheme 1. . . Scheme 2”

16 removed all contractions, such as “can’t” or “couldn’t”? and changed them to “cannot” or “could not”?

17 checked the columns of numbers in your Tables and made all the decimal points in the same column line up vertically?

18 put a noun right after “this” or “these” every time you use one of those words?

19 moved “however” away from the beginning of sentences to a spot later in a sentence where it sounds better?

20 made all your abbreviations the same?

NOT 48 hrs 3 hours 2 h five h

BUT 48 h 3 h 2 h 5 h

21 used strong definite words?

NOT “We have been interested in”

BUT “We have focused on”

22 used parallel wording where you lead your readers to expect it?

NOT “not only for fabricating. .but also for reduction”

BUT “not only for fabricating. . .but also for reducing. . .”

OR “not only for the formation of. . .but also for the elimination of. . “

Don’t mix words ending in “ing” with “ion” words.

23 proofread your paper 3 times? At least!

24 read your paper aloud to yourself?

25 Did you use “later” when you meant “latter”?

26 looked for fruit in your figure captions?

A fig is a fruit you can eat, but “Fig.” is the correct abbreviation for “Figure.” Don’t forget the period in the abbreviation.

27 proofread your references? Made sure that a journal title, no matter how often it occurs, is always spelled the same way?

28 followed the reference format for the journal you are submitting to?

深度学习是机器学习的一个子领域,它基于人工神经网络的研究,特别是利用多层次的神经网络来进行学习和模式识别。深度学习模型能够学习数据的高层次特征,这些特征对于图像和语音识别、自然语言处理、医学图像分析等应用至关重要。以下是深度学习的一些关键概念和组成部分: 1. **神经网络(Neural Networks)**:深度学习的基础是人工神经网络,它是由多个层组成的网络结构,包括输入层、隐藏层和输出层。每个层由多个神经元组成,神经元之间通过权重连接。 2. **馈神经网络(Feedforward Neural Networks)**:这是最常见的神经网络类型,信息从输入层流向隐藏层,最终到达输出层。 3. **卷积神经网络(Convolutional Neural Networks, CNNs)**:这种网络特别适合处理具有网格结构的数据,如图像。它们使用卷积层来提取图像的特征。 4. **循环神经网络(Recurrent Neural Networks, RNNs)**:这种网络能够处理序列数据,如时间序列或自然语言,因为它们具有记忆功能,能够捕捉数据中的时间依赖性。 5. **长短期记忆网络(Long Short-Term Memory, LSTM)**:LSTM 是一种特殊的 RNN,它能够学习长期依赖关系,非常适合复杂的序列预测任务。 6. **生成对抗网络(Generative Adversarial Networks, GANs)**:由两个网络组成,一个生成器和一个判别器,它们相互竞争,生成器生成数据,判别器评估数据的真实性。 7. **深度学习框架**:如 TensorFlow、Keras、PyTorch 等,这些框架提供了构建、训练和部署深度学习模型的工具和库。 8. **激活函数(Activation Functions)**:如 ReLU、Sigmoid、Tanh 等,它们在神经网络中用于添加非线性,使得网络能够学习复杂的函数。 9. **损失函数(Loss Functions)**:用于评估模型的预测与真实值之间的差异,常见的损失函数包括均方误差(MSE)、交叉熵(Cross-Entropy)等。 10. **优化算法(Optimization Algorithms)**:如梯度下降(Gradient Descent)、随机梯度下降(SGD)、Adam 等,用于更新网络权重,以最小化损失函数。 11. **正则化(Regularization)**:技术如 Dropout、L1/L2 正则化等,用于防止模型过拟合。 12. **迁移学习(Transfer Learning)**:利用在一个任务上训练好的模型来提高另一个相关任务的性能。 深度学习在许多领域都取得了显著的成就,但它也面临着一些挑战,如对大量数据的依赖、模型的解释性差、计算资源消耗大等。研究人员正在不断探索新的方法来解决这些问题。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值