## Can you invert convolution?

In mathematics, **deconvolution is the operation inverse to convolution**. Both operations are used in signal processing and image processing. For example, it may be possible to recover the original signal after a filter (convolution) by using a deconvolution method with a certain degree of accuracy.

**What is the inverse of image convolution?**

Convolution is a technique used to enhance specific characteristics of an image, while **deconvolution** is its inverse process.

**Why do we reverse convolution?**

If you picture a sliding function over a stationary function (which is what convolution does), if the sliding one is not reversed (either of them, actually), **you will convolute the present time for the stationary function with the future time of the sliding function, which disrespects order of time events**.

**What is the inverse of discrete convolution?**

Computing the inverse of the convolution operation is known as **deconvolution**.

**Is the convolution operation irreversible?**

But in general, convolution of functions is almost a ring (there's no exact identity element). The linear space of compactly supported distributions forms an actual ring under convolution, and so it has a group of units. These are distributions whose **convolutions are always reversible**.

**What is the inverse problem of convolution?**

The inverse problem related to convolution is called deconvolution. The observed data is interpreted as **clean signal convolved with a kernel and corrupted with random noise**. The goal of deconvolution is to reconstruct the clean signal from the noisy data.

**What is the drawback of convolution?**

The biggest disadvantage of Convolution is the **inherent locality constraints**. Convolution is limited in its ability to extract visual patterns across different spatial positions. The receptive field of each pixel is narrow making it very hard to model long-range multihop dependencies amongst the pixels.

**What is the purpose of deconvolution?**

Deconvolution is a computationally intensive image processing technique **to reassign light back to its correct plane of focus, producing images with higher contrast and resolution**.

**Why do we pool after convolution?**

In convolutional neural networks (CNNs), the pooling layer is a common type of layer that is typically added after convolutional layers. The pooling layer is used **to reduce the spatial dimensions (i.e., the width and height) of the feature maps, while preserving the depth (i.e., the number of channels)**.

**What is image deconvolution?**

Image deconvolution is **an image-processing technique designed to remove blur or enhance contrast and resolution**. Historically, its application was limited to the enhancement of widefield images and it was considered unnecessary in confocal microscopy.

## Is convolution linear or nonlinear?

Typical convolutional layers are **linear systems**, hence their expressiveness is limited. To overcome this, various non-linearities have been used as activation functions inside CNNs, while also many pooling strategies have been applied.

**Is convolution just multiplication?**

Algebraically, **convolution is the same operation as multiplying polynomials whose coefficients are the elements of u and v** . w ( k ) = ∑ j u ( j ) v ( k − j + 1 ) .

**Is convolution always continuous?**

In this sense f is only defined a.e., yet **the convolution f ∗ g is a continuous function that is defined everywhere**. In particular, changing f on a set of measure zero has no effect on value of (f ∗ g)(x) = ∫ f(y) g(x − y) dy.

**What happens if you convolve something with itself?**

Answer and Explanation: The convolution of a function by itself means **the squaring of that function in the frequency domain**. For example, Let be a density function, in that case, k ( x ) = ( h ∗ h ) ( x ) is the squaring of in the frequency domain. This type of convolution is termed as the Power Spectrum.

**Is kernel flipped in convolution?**

In Convolution operation, **the kernel is first flipped by an angle of 180 degrees and is then applied to the image**. The fundamental property of convolution is that convolving a kernel with a discrete unit impulse yields a copy of the kernel at the location of the impulse.

**What are the alternatives to convolution?**

**OctConv** can simply replace a standard convolution in neural networks without requiring any other network architecture adjustments.

**Does convolution have bias?**

Parameters of a Convolutional Layer

**There is one bias for each output channel**. Each bias is added to every element in that output channel.

**What is asymmetric convolution?**

asymmetric convolutions work by **taking the x and y axes of the image separately**. For example performing a convolution with an (n×1) kernel before one with a (1×n) kernel.

**What is the opposite of a convolutional neural network?**

**Deconvolutional networks** are convolutional neural networks (CNN) that work in a reversed process.

**Why does CNN use convolution?**

Convolution is a mathematical operation that allows the merging of two sets of information. In the case of CNN, convolution is applied to the input data **to filter the information and produce a feature map**. This filter is also called a kernel, or feature detector, and its dimensions can be, for example, 3x3.

## What are the weaknesses of convolutional neural networks?

However, CNNs also have some drawbacks that limit their performance and applicability. One of the main disadvantages of CNNs is that **they require a large amount of labeled data to train effectively**, which can be costly and time-consuming to obtain and annotate.

**Why CNN is better than other algorithms?**

Benefits of Using CNNs for Machine and Deep Learning

The CNN architecture is especially useful for image recognition and image classification, as well as other computer vision tasks because **they can process large amounts of data and produce highly accurate predictions**.

**Why is deconvolution hard?**

Deconvolution aims to reverse this process and recover the original image from the observed image and the PSF. However, in many situations, **the PSF is unknown or only partially known**, which makes deconvolution more challenging and prone to errors.

**What's the difference between convolution and deconvolution?**

In DeConvolution, the Feature map gets converted into an Image. Converting Convoluted image into Original Image is DeConvolution. **Convolution adds up the information spread in various pixels to one pixel, whereas DeConvolution spreads the information present in one pixel to various pixels**.

**What are the two types of deconvolution?**

The result of both methods (a) and (b) is an improvement in image contrast, making it easier to resolve and distinguish features in the image. There are two main types of deconvolution algorithm: **deblurring and restorative**.

**Does pooling prevent overfitting?**

Besides, pooling provides the ability to learn invariant features and also **acts as a regularizer to further reduce the problem of overfitting**.

**Which is better average pooling or max pooling?**

Average pooling method smooths out the image and hence the sharp features may not be identified when this pooling method is used. **Max pooling selects the brighter pixels from the image**. It is useful when the background of the image is dark and we are interested in only the lighter pixels of the image.

**What is the difference between convolution and pooling?**

How is convolution different from pooling? The significant difference is that **a convolution layer extracts features from the data matrix, whereas the pooling layer only downsamples the data matrix**.

**Does deconvolution improve resolution?**

In fact, **deconvolution algorithms alone can reduce the size of a sub-resolution fluorescent bead full width at half maximum (FWHM) to achieve a measured 2x increase in resolution over widefield**.

**What is the best deconvolution algorithm?**

**The CMLE** is the most general deconvolution algorithm available in Huygens, that provides good results for a variety of imaging conditions. You can optimize your CMLE results by fine-tuning the Acuity setting or, if needed, you can adjust the SNR value or number of iterations.

## Is deconvolution same as upsampling?

Transposed convolution is also known as Deconvolution which is not appropriate as deconvolution implies removing the effect of convolution which we are not aiming to achieve. **It is also known as upsampled convolution** which is intuitive to the task it is used to perform, i.e upsample the input feature map.

**Is convolution always commutative?**

**The operation of convolution is commutative**.

**Why DFT does not support linear convolution?**

Therefore when doing a DFT operation, **theres is an implicit alteration to your signal from being finite, to being periodic, even if your signal is not periodic**. This periodicity of the signal leads to the need of using convolution in a circular manner.

**Why is nonlinearity needed in CNN?**

This nonlinearity **allows neural networks to develop complex representations and functions based on the inputs that would not be possible with a simple linear regression model**.

**Is the Fourier transform a convolution?**

We've just shown that **the Fourier Transform of the convolution of two functions is simply the product of the Fourier Transforms of the functions**. This means that for linear, time-invariant systems, where the input/output relationship is described by a convolution, you can avoid convolution by using Fourier Transforms.

**Is discrete time convolution possible?**

1. Is discrete time convolution possible? Explanation: Yes, like continuous time convolution discrete time convolution is also possible with the same phenomena except that it is discrete and superimposition occurs only in those time interval in which signal is present.

**Can you divide a convolution?**

**Yes, you can split them up**. Convolution is a linear process, which means that superposition holds. Thus, you can break up any convolution kernel k into multiple parts (k1,k2,... kN) such that k=∑ki.

**What is a real life example of convolution?**

One of the real life applications of convolution is **seismic signals for oil exploration**. Here a perturbation is produced in the surface of the area to be analized. The signal travel underground producing reflexions at each layer. This reflexions are measured in the surface through a sensors network.

**Why is convolution better than fully connected?**

**Convolutions are not densely connected, not all input nodes affect all output nodes**. This gives convolutional layers more flexibility in learning. Moreover, the number of weights per layer is a lot smaller, which helps a lot with high-dimensional inputs such as image data.

**Why are convolution kernels odd?**

In practice, an odd-sized kernel is preferred because **it makes it possible to validate the coordinates of the origin within the kernel**. This reduces the distortion between layers as the previous layer pixels would be symmetrical around the output pixel.

## Does convolution have an identity?

Identity **The delta function is the identity for convolution**. Convolving a signal with the delta function leaves the signal unchanged. This is the goal of systems that transmit or store signals.

**What is the difference between convolution and correlation?**

**Convolution is the calculation of the area under the product of two signals in LTI systems where as correlation is measurement of similarity between two signals**. Correlation is measurement of the similarity between two signals/sequences. Convolution is measurement of effect of one signal on the other signal.

**Is multiplication distributive over convolution?**

**Binary operations like multiply and convolution are both somehow of multiplicative structure**. Thus, I suspect there is no generic distributivity (apart for Boole or Boolean algebras, where the restriction on allowed values shrinks the problem, see below).

**What is twisted convolution?**

Remark 4 Twisted convolution is actually just **a component of ordinary convolution, but not on the original group ; instead it relates to convolution on a Heisenberg group extension of this group**. More specifically, define the Heisenberg group to be the set of pairs with group law. and inverse operation.

**Is convolution like dot product?**

**The convolution is calculated by reversing the second vector and sliding it across the first vector**. The dot product of the two vectors is calculated at each point as the second vector is slid across the first vector. The dot products are collected in a third vector which is the convolution of the two vectors.

**Is transposed convolution inverse of convolution?**

**It is the inverse of the multivariate convolutional function**. On the other hand, a transposed convolutional layer only reconstructs the spatial dimensions of the input.

**What is opposite of convolution?**

**Deconvolution, or polynomial division**, is the inverse operation of convolution. Deconvolution is useful in recovering the input to a known filter, given the filtered output. This method is very sensitive to noise in the coefficients, however, so use caution in applying it. The syntax for deconv is. [q,r] = deconv(b,a)

**What is an atrous convolution?**

Put simply, atrous convolution is **akin to the standard convolution except that the weights of an atrous convolution kernel are spaced r locations apart**, i.e., the kernel of dilated convolution layers are sparse.

**Which algorithm is better than CNN?**

There are two major alternatives to CNN (Convolutional Neural Network) namely: **Graph Neural Networks**. Capsule Neural Network.

**What is inverse of an image?**

inverse image (plural inverse images) **The set of points that map to a given point (or set of points) under a specified function**.

## What does it mean to inverse an image?

Color inversion is **a photo effect that flips all colors to their opposite hue on the color wheel**. For example, white becomes black, green becomes magenta, and blue becomes orange. The interest in color-inverted photography started with analog photography and film negatives.

**What is the invert function in image?**

CSS invert() Function

The invert() function is an inbuilt function that is **used to apply a filter to the image to set the invert of the color of the sample image**.

**What is the inverse image problem?**

Inverse problems involve **estimating parameters or data from inadequate observations**; the observations are often noisy and contain incomplete information about the target parameter or data due to physical limitations of the measurement devices.

**When can an image be inverted?**

* **If the image is real in the case of lenses**, it will always be inverted. Also in case of lenses if the image distance is positive hence the image is inverted. These images will form on the opposite side of the lens. * If the image is virtual in the case of mirrors, it will always be upright.

**Can a real image be inverted?**

Real images are always located behind the mirror. **Real images can be either upright or inverted**. Real images can be magnified in size, reduced in size or the same size as the object. Real images can be formed by concave, convex and plane mirrors.

**How do you make an image inverse?**

**How to invert an image in 4 steps**

- Upload an image in PNG or JPG format or drag and drop it in the editor.
- Click on the Image Effects & Filters tool on the top left toolbar.
- Scroll down to Filters and select Invert to apply it to your photo.

**How do I mirror an image?**

Mirror your image

**Place the duplicate photo side by side with your original image.** Then select the duplicate to show the menu bar. Click Flip then choose between Flip Horizontal or Flip Vertical depending on your needs.

**What is an example of an inverted image?**

**Imagine a tree upside down**. The position of the tree is said to be inverted. Thus images which appear to be upside down are inverted images. In other words its an image rotated at 180 degrees from the line of sight between the observer and the object.

**What is the opposite of an inverted image?**

In an **erect image**, directions are the same as those in the object, in contrast to an inverted image.

**Is an inverse function a mirror image?**

i.e **The graph of an inverse function can be obtained from thecorresponding graph of original function as a mirror image** (i.e., reflection) along the line y = x. This can be understood visually too from the graph of sin x and sin^{-}^{1}x drawn on the same set of axes.

## What is the difference between image and inverse function?

The biggest difference between a preimage and the inverse function is that **the preimage is a subset of the domain.** **The inverse (if it exists) is a function between two sets**. In that sense they are two very different animals. A set and a function are completely different objects.

**What is inverse mapping in image processing?**

An inverse mapping function **maps output pixels to the corresponding input area**. In other words, the function allows you to compute the distorted area on the input corresponding to a rectangular output pixel.

## References

- https://math.stackexchange.com/questions/2232020/why-do-we-need-to-reverse-a-function-in-the-convolution-operaton
- https://pixelied.com/features/photo-filters/invert-photo
- https://encyclopediaofmath.org/wiki/Convolution_of_functions
- https://iq.opengenus.org/alternatives-to-cnn/
- https://www.purdue.edu/freeform/ervibrations/chapter-iv-animations/convolution-integral-interpretation/
- https://dsp.stackexchange.com/questions/35155/why-is-circular-convolution-used-in-dsp-why-not-linear-convolution
- https://www.thesaurus.com/browse/convolution
- https://www.projectpro.io/recipes/what-is-role-of-max-pooling-operation-neural-network
- https://en.wikipedia.org/wiki/Erect_image
- https://betterexplained.com/articles/intuitive-convolution/
- https://www.mathworks.com/help/fixedpoint/ref/conv.html
- https://medium.com/@bdhuma/which-pooling-method-is-better-maxpooling-vs-minpooling-vs-average-pooling-95fb03f45a9
- https://www.knowledgehut.com/blog/data-science/pooling-layer
- https://www.sanfoundry.com/signals-systems-questions-answers-online-test/
- https://www.analyticsvidhya.com/blog/2021/04/invert-the-characteristics-of-convolution-with-involution/
- https://www.analog.com/media/en/technical-documentation/dsp-book/dsp_book_Ch7.pdf
- https://www.laserfocusworld.com/optics/article/16556298/microscopyimage-processing-a-deconvolution-revolution-for-confocal-image-enhancement
- https://towardsdatascience.com/a-primer-on-atrous-convolutions-and-depth-wise-separable-convolutions-443b106919f5
- https://math.stackexchange.com/questions/373455/is-the-convolution-an-invertible-operation
- https://www.vedantu.com/question-answer/tell-if-an-image-is-upright-or-inverted-class-12-physics-jee-main-637366c25c4463725f7d33f6
- https://brainly.in/question/6788383
- https://prvnk10.medium.com/the-convolution-operation-48d72a382f5a
- https://stats.stackexchange.com/questions/486376/why-do-we-even-need-max-pooling-layers
- https://datagen.tech/guides/computer-vision/cnn-convolutional-neural-network/
- https://ai.stackexchange.com/questions/7294/optimizing-max-pooling-algorithm
- https://www.sanfoundry.com/signals-systems-questions-answers-discrete-time-convolution-1/
- https://byjus.com/question-answer/the-graph-of-an-inverse-function-can-be-obtained-from-the-corresponding-graph-of-original/
- https://elec3004.uqcloud.net/2013/lectures/Convolution%20(Sec%201.4%20from%20Hayes%20DSP).pdf
- https://ai.stackexchange.com/questions/13738/what-is-the-difference-between-asymmetric-and-depthwise-separable-convolution
- https://machinelearningmastery.com/pooling-layers-for-convolutional-neural-networks/
- https://www.researchgate.net/figure/Toy-example-illustrating-the-drawbacks-of-max-pooling-and-average-pooling_fig2_300020038
- https://aaweg-i.medium.com/computer-vision-2-maxpooling-and-dropouts-f89cfc5fa412
- https://leesael.github.io/talk/BIML2017/part2-2.pdf
- https://eng.libretexts.org/Bookshelves/Electrical_Engineering/Signal_Processing_and_Modeling/Signals_and_Systems_(Baraniuk_et_al.)/03%3A_Time_Domain_Analysis_of_Continuous_Time_Systems/3.04%3A_Properties_of_Continuous_Time_Convolution
- https://www.calbanktrust.com/personal/community/blog/backyard-pool-pros-cons/
- https://www.olympus-lifescience.com/en/discovery/the-race-to-super-resolution-microscopy-is-deconvolution-enough/
- https://www.sfu.ca/~truax/conv.html
- https://www.merriam-webster.com/dictionary/convolution
- https://www.canva.com/features/mirror-image/
- https://medium.com/geekculture/max-pooling-why-use-it-and-its-advantages-5807a0190459
- https://www.geeksforgeeks.org/convolutional-block-for-image-recognition/
- https://medium.com/analytics-vidhya/understanding-convolution-operations-in-cnn-1914045816d4
- https://towardsdatascience.com/understanding-convolutions-and-pooling-in-neural-networks-a-simple-explanation-885a2d78f211
- https://stackoverflow.com/questions/54359135/should-i-use-conv-3x3-in-last-convolutional-layer-for-semantic-segmentation
- https://link.springer.com/article/10.1007/s00500-021-05783-5
- https://study.com/academy/lesson/convolution-theorem-application-examples.html
- https://en.wikipedia.org/wiki/Convolution
- https://medium.com/@marsxiang/convolutions-transposed-and-deconvolution-6430c358a5b6
- https://saturncloud.io/blog/where-to-add-dropout-in-neural-network/
- https://paperswithcode.com/method/convolution
- https://iq.opengenus.org/output-size-of-convolution/
- https://www.sciencedirect.com/topics/mathematics/convolution-theorem
- https://dsp.stackexchange.com/questions/59221/is-convolution-distributive-over-multiplication
- https://en.wikipedia.org/wiki/Convolution_theorem
- https://www.tutorialspoint.com/digital_signal_processing/dsp_operations_on_signals_convolution.htm
- https://www.linkedin.com/advice/1/how-do-you-balance-trade-off-between-regularization
- https://weinman.cs.grinnell.edu/courses/CSC161/2021F/homework/convolution.shtml
- https://studies.helsinki.fi/courses/cu/hy-CU-117630717-2021-08-01
- https://math.fel.cvut.cz/en/mt/txta/1/txe3aa1d.htm
- https://www.quora.com/What-is-the-difference-between-convolution-and-multiplication-of-two-signals
- https://www.mathworks.com/help/signal/ug/deconvolution.html
- https://math.stackexchange.com/questions/3286150/understanding-the-difference-between-pre-image-and-inverse
- https://www.codingninjas.com/studio/library/convolution-layer-padding-stride-and-pooling-in-cnn
- https://www.mathworks.com/matlabcentral/answers/709118-create-a-7x7-convolution-kernel-which-has-an-equivalent-effect-to-three-passes-with-a-3x3-mean-filte
- https://waterprogramming.wordpress.com/2018/09/04/implementation-of-the-moving-average-filter-using-convolution/
- https://towardsdatascience.com/a-comprehensive-introduction-to-different-types-of-convolutions-in-deep-learning-669281e58215
- http://www.microscopist.co.uk/essential-techniques/deconvolution/
- https://www.websupergoo.com/helpie/source/2-effects/convolution.htm
- https://www.geeksforgeeks.org/cnn-introduction-to-padding/
- https://towardsdatascience.com/beginners-guide-to-understanding-convolutional-neural-networks-ae9ed58bb17d
- https://www.nandantechnicals.com/2021/06/what-is-linear-convolution-and.html
- https://www.statlect.com/glossary/convolutions
- https://medium.com/analytics-vidhya/the-perfect-fit-for-a-dnn-596954c9ea39
- https://towardsdatascience.com/why-convolve-understanding-convolution-and-feature-extraction-in-deep-networks-ee45d1fdd17c
- https://datascience.stackexchange.com/questions/44518/can-pooling-ever-increase-accuracy-in-convolutional-neural-networks
- https://www.sciencedirect.com/topics/computer-science/convolution-operation
- https://www.sciencedirect.com/topics/neuroscience/deconvolution
- https://class.ece.uw.edu/235dl/EE235/Project/lesson17/lesson17.html
- https://www.researchgate.net/post/Difference_between_convolution_and_correlation
- https://towardsai.net/p/l/introduction-to-pooling-layers-in-cnn
- https://deeplizard.com/learn/video/ZjM_XQa5s6s
- https://stackoverflow.com/questions/43485361/whats-the-difference-between-conv-layer-and-pooling-layer-in-cnn
- https://www.linkedin.com/advice/0/what-advantages-disadvantages-using-convolutional
- https://www.mathworks.com/discovery/convolutional-neural-network-matlab.html
- https://www.researchgate.net/post/What-are-the-real-life-applications-of-correlation-and-convolution
- https://mathworld.wolfram.com/Convolution.html
- https://dsp.stackexchange.com/questions/4723/what-is-the-physical-meaning-of-the-convolution-of-two-signals
- https://stats.stackexchange.com/questions/413275/why-do-we-use-the-average-pooling-layers-instead-of-max-pooling-layers-in-the-de
- https://www.cs.toronto.edu/~lczhang/360/lec/w04/convnet.html
- https://stats.stackexchange.com/questions/288261/why-is-max-pooling-necessary-in-convolutional-neural-networks
- https://solr.apache.org/guide/7_6/dsp.html
- https://staff.fnwi.uva.nl/r.vandenboomgaard/IPCV20162017/LectureNotes/IP/LocalOperators/convolutionExamples.html
- https://en.wiktionary.org/wiki/inverse_image
- https://www.baeldung.com/cs/neural-networks-conv-fc-layers
- https://www.tutorialspoint.com/what-is-convolution-in-signals-and-systems
- https://homework.study.com/explanation/what-happens-when-you-convolve-a-function-by-itself.html
- https://svi.nl/Deconvolution-Algorithms
- https://www.techtarget.com/searchenterpriseai/definition/deconvolutional-networks-deconvolutional-neural-networks
- https://towardsdatascience.com/convolutional-layers-vs-fully-connected-layers-364f05ab460b
- https://towardsdatascience.com/convolution-vs-correlation-af868b6b4fb5
- https://www.linkedin.com/advice/0/what-best-methods-deconvolving-images-unknown
- https://www.analog.com/media/en/technical-documentation/dsp-book/dsp_book_Ch6.pdf
- https://towardsdatascience.com/understanding-and-calculating-the-number-of-parameters-in-convolution-neural-networks-cnns-fc88790d530d
- https://www.cyberlink.com/blog/photo-editing-best-software/1105/how-to-invert-colors-in-an-image
- https://datascience.stackexchange.com/questions/46858/how-does-max-pooling-of-size-2x2-change-the-size-of-receptive-field-in-cnn
- https://saturncloud.io/blog/understanding-the-difference-between-flatten-and-globalaveragepooling2d-in-keras/
- https://sciendo.com/pdf/10.2478/fcds-2019-0016
- https://iq.opengenus.org/cnn-questions/
- https://www.mathworks.com/matlabcentral/answers/460828-multiplication-in-frequency-domain-equals-convolution-in-time-domain-mismatch
- https://towardsdatascience.com/covolutional-neural-network-cb0883dd6529
- https://www.devron.ai/kbase/how-to-choose-the-optimal-kernel-size
- https://heil.math.gatech.edu/6338/summer08/section4c_convolve.pdf
- https://www.simplilearn.com/tutorials/deep-learning-tutorial/convolutional-neural-network
- https://towardsdatascience.com/transposed-convolution-demystified-84ca81b4baba
- https://towardsdatascience.com/discuss-the-dog-image-classification-project-with-cnn-bc5c90ee4fec
- https://arxiv.org/abs/1708.07038
- https://towardsdatascience.com/convolution-a-journey-through-a-familiar-operators-deeper-roots-2e3311f23379
- https://www.geeksforgeeks.org/cnn-introduction-to-pooling-layer/
- https://www.analog.com/media/en/technical-documentation/dsp-book/dsp_book_ch6.pdf
- https://stackoverflow.com/questions/3323063/what-is-an-inverse-mapping-function-in-image-processing
- https://dsp.stackexchange.com/questions/6298/breaking-a-convolution-into-smaller-pieces
- https://towardsdatascience.com/how-to-reduce-training-parameters-in-cnns-while-keeping-accuracy-99-a213034a9777
- https://terrytao.wordpress.com/tag/twisted-convolution/
- https://www.physicsclassroom.com/reviews/Reflection-and-Mirrors/Reflection-and-Mirrors-Review-Answers-2
- https://www.baeldung.com/cs/ml-relu-dropout-layers
- https://www.sciencedirect.com/topics/engineering/convolutional-layer
- https://www.geeksforgeeks.org/css-invert-function/
- https://www.upgrad.com/blog/basic-cnn-architecture/
- https://mathoverflow.net/questions/3455/do-convolution-and-multiplication-satisfy-any-nontrivial-algebraic-identities
- https://medium.com/@iamvarman/how-to-calculate-the-number-of-parameters-in-the-cnn-5bd55364d7ca
- https://scholarship.rice.edu/handle/1911/18675
- https://machinelearningmastery.com/convolutional-layers-for-deep-learning-neural-networks/
- https://www.quora.com/Why-is-convolution-in-time-domain-equivalent-to-frequency-multiplication
- https://ai.meta.com/blog/octconv-a-flexible-efficient-alternative-to-standard-convolution/
- https://www.sciencedirect.com/topics/mathematics/pooling-layer
- https://serokell.io/blog/introduction-to-convolutional-neural-networks
- https://stackoverflow.com/questions/59717290/does-maxpooling-reduce-overfitting
- https://www.mathworks.com/help/matlab/ref/conv.html
- https://medium.com/geekculture/deconvolution-and-how-it-works-and-its-importance-80ce8f137e1b
- https://blog.francium.tech/machine-learning-convolution-for-image-processing-42623c8dbec0
- https://accountlearning.com/advantages-and-disadvantages-of-pool-organizations/
- https://www.linkedin.com/pulse/why-activation-function-used-neural-network-mayur-ingole