TargetLink is a software for automatic code generation, based on a subset of Simulink/Stateflow models, produced by dSPACE GmbH. TargetLink requires an existing MATLAB/Simulink model to work on. TargetLink generates both ANSI-C and production code optimized for specific processors. It also supports the generation of AUTOSAR-compliant code for software components for the automotive sector. The management of all relevant information for code generation takes place in a central data container, called the Data Dictionary. Testing of the generated code is implemented in Simulink, which is also used for the specification of the underlying simulation models. TargetLink supports three simulation modes to test the generated code: Model-in-the-loop simulation (MIL): this mode allows the model design to be checked. An MIL simulation is also known as a floating-point simulation, since the variables are typically floating-point variables. Software-in-the-loop (SIL): the simulation is based on the execution of generated code, which runs on a PC system. The variables are typically plain or fixed point numbers. Processor-in-the-loop (PIL): in a PIL simulation, the generated code runs on the target hardware or on an evaluation board. So-called real-time frames are included, making it possible to transfer the simulation results as well as memory consumption and runtime information to the PC. The Motor Industry Software Reliability Association (MISRA) published official MISRA modeling guidelines for TargetLink in late 2007, which are particularly important for functional safety of safety-critical applications. In 2009, TÜV SÜD certified TargetLink for use during the development of safety-critical systems to ISO DIS 26262 and IEC 61508.
Cloudlet
A cloudlet is a mobility-enhanced small-scale cloud datacenter that is located at the edge of the Internet. The main purpose of the cloudlet is supporting resource-intensive and interactive mobile applications by providing powerful computing resources to mobile devices with lower latency. It is a new architectural element that extends today's cloud computing infrastructure. It represents the middle tier of a 3-tier hierarchy: mobile device - cloudlet - cloud. A cloudlet can be viewed as a data center in a box whose goal is to bring the cloud closer. The cloudlet term was first coined by M. Satyanarayanan, Victor Bahl, Ramón Cáceres, and Nigel Davies, and a prototype implementation is developed by Carnegie Mellon University as a research project. The concept of cloudlet is also known as follow me cloud, and mobile micro-cloud. == Motivation == Many mobile services split the application into a front-end client program and a back-end server program following the traditional client-server model. The front-end mobile application offloads its functionality to the back-end servers for various reasons such as speeding up processing. With the advent of cloud computing, the back-end server is typically hosted at the cloud datacenter. Though the use of a cloud datacenter offers various benefits such as scalability and elasticity, its consolidation and centralization lead to a large separation between a mobile device and its associated datacenter. End-to-end communication then involves many network hops and results in high latencies and low bandwidth. For the reasons of latency, some emerging mobile applications require cloud offload infrastructure to be close to the mobile device to achieve low response time. In the ideal case, it is just one wireless hop away. For example, the offload infrastructure could be located in a cellular base station or it could be LAN-connected to a set of Wi-Fi base stations. The individual elements of this offload infrastructure are referred to as cloudlets. == Applications == Cloudlets aim to support mobile applications that are both resource-intensive and interactive. Augmented reality applications that use head-tracked systems require end-to-end latencies of less than 16 ms. Cloud games with remote rendering also require low latencies and high bandwidth. Wearable cognitive assistance systems combine devices such as Google Glass with cloud-based processing to guide users through complex tasks. This futuristic genre of applications is characterized as “astonishingly transformative” by the report of the 2013 NSF Workshop on Future Directions in Wireless Networking. These applications use cloud resources in the critical path of real-time user interaction. Consequently, they cannot tolerate end-to-end operation latencies of more than a few tens of milliseconds. Apple Siri and Google Now which perform compute-intensive speech recognition in the cloud, are further examples in this emerging space. == Cloudlet vs Cloud == There is significant overlap in the requirements for cloud and cloudlet. At both levels, there is the need for: (a) strong isolation between untrusted user-level computations; (b) mechanisms for authentication, access control, and metering; (c) dynamic resource allocation for user-level computations; and, (d) the ability to support a very wide range of user-level computations, with minimal restrictions on their process structure, programming languages or operating systems. At a cloud datacenter, these requirements are met today using the virtual machine (VM) abstraction. For the same reasons they are used in cloud computing today, VMs are used as an abstraction for cloudlets. Meanwhile, there are a few but important differentiators between cloud and cloudlet. === Rapid provisioning === Different from cloud data centers that are optimized for launching existing VM images in their storage tier, cloudlets need to be much more agile in their provisioning. Their association with mobile devices is highly dynamic, with considerable churn due to user mobility. A user from far away may unexpectedly show up at a cloudlet (e.g., if he just got off an international flight) and try to use it for an application such as a personalized language translator. For that user, the provisioning delay before he is able to use the application impacts usability. === VM handoff across cloudlets === If a mobile device user moves away from the cloudlet he is currently using, the interactive response will degrade as the logical network distance increases. To address this effect of user mobility, the offloaded services on the first cloudlet need to be transferred to the second cloudlet maintaining end-to-end network quality. This resembles live migration in cloud computing but differs considerably in a sense that the VM handoff happens in Wide Area Network (WAN). == OpenStack++ == Since the cloudlet model requires reconfiguration or additional deployment of hardware/software, it is important to provide a systematic way to incentivise the deployment. However, it can face a classic bootstrapping problem. Cloudlets need practical applications to incentivize cloudlet deployment. However, developers cannot heavily rely on cloudlet infrastructure until it is widely deployed. To break this deadlock and bootstrap the cloudlet deployment, researchers at Carnegie Mellon University proposed OpenStack++ that extends OpenStack to leverage its open ecosystem. OpenStack++ provides a set of cloudlet-specific APIs as OpenStack extensions. == Commercial implementations and standardization effort == By 2015 cloudlet based applications were commercially available. In 2017 the National Institute of Standards and Technology published draft standards for fog computing in which cloudlets were defined as nodes on the fog architecture.
Talking Angela
Talking Angela is a mobile game (formerly a chatbot), developed by Slovenian studio Outfit7 as part of the Talking Tom & Friends series. It was released on 13 November 2012 and December 2012 for iPhone, iPod and iPad, January 2013 for Android, and January 2014 for Google Play. The game's successor, the My Talking Angela game, was released in December 2014. The game takes place in a café in Paris and allows players to interact with Angela, an anthropomorphic white cat in different ways. Players can use coins to purchase makeup, accessories and items, as well as drinks that will trigger different visual effects. The fortune cookie button causes Angela to read out a fortune cookie, while the bird icon will prompt birds to fly around the screen, or have Angela feed them. Players can also pet or poke Angela, as well the café's sign. Prior to their removal, the game featured a chat system and a camera button. Users can engage in conversations with Angela, ask for quizzes or initiate a short snippet of the song "That's Falling In Love". If the player was to type in "Who is an idiot?", Angela would respond with a random swear word. Additionally, inquiring Angela about sexual topics would cause her to reply with "Do you want to talk about sex?", though she will quickly change the topic regardless of what the player writes next. A hoax claiming that Angela's eyes were hidden cameras that enabled hackers or paedophiles to watch children was spread. Despite the claims, Snopes and The Guardian found no evidence. Due to the hoax, Angela received a blue dress, as well as an altered eye asset with a different reflection, and later the chat and camera functions were removed altogether. == Hoaxes == In February 2014, Talking Angela was the subject of an Internet hoax alleging that the application was a front for child predators to exploit children. The rumor, which was widely circulated on Facebook and various websites claiming to be dedicated to parenting, claims that a sinister sexual predator or hacker, asked children for private personal information using the game's text-chat feature. Other versions of the rumour even attributed the disappearance of a child to the game; one news report claimed that a seven year old boy disappeared after downloading the app. Another variation included that it was run by a paedophile ring, citing a man that could be seen in Angela's eyes. The app's developers, Outfit7, later gave a statement refuting the hoaxes. The hoax was eventually debunked by Snopes, a fact-checking website. The site's owners, Barbara and David Mikkelson, reported that they had tried to "prompt" it to give responses asking for private information, but were unsuccessful, even when asking it explicitly sexual questions. While it is true that, in the game with child mode off, Angela does ask for the user's name, age and personal preferences to determine conversation topics, Outfit7 has said that this information is all "anonymized" and all personal information is removed from it. It is also impossible for a person to take control of what Angela says in the game, since the game is based on chatbot software. When the mode was turned on, the chat feature was disabled, meaning no personal questions could be asked. In 2015, the hoax was revived on Facebook, which prompted online security company Sophos and The Guardian to debunk it again. Sophos employee Paul Ducklin wrote that the message being posted on Facebook promoting the hoax was "close to 600 rambling, repetitious words, despite claiming at the start that it didn't have words to describe the situation. It's ill-written, and borders on being illiterate and incomprehensible." Bruce Wilcox, one of the game's programmers, attributed the hoax's popularity to the fact that the chatbot program in Talking Angela aimed to sound realistic. Concern was raised that the game's child mode may have been too easy for children to turn off. It allowed them to purchase "coins", premium currency in the game, via iTunes, and enabled the chat feature. While not "connecting your children to paedophiles", this still raised concerns according to The Guardian. === Impact === The scare significantly boosted the game's popularity, and was credited with helping the app enter the top 10 free iPhone apps soon after the hoax became widely known in February 2015,In the truth the reason there is a man in Angela’s eyes is because of pareidoila, the ability to see through diamonds and other minerals and water bodies and shiny objects,which is the reason why players notice a man in her eyes,The truth is that being Angela’s eyes simply serve as a reflective surface,Because of the low quality of this reflection the reflection was mistaken for a humanoid figure. oref>Smith, Josh (19 February 2014). "Talking Angela App Scare Skyrockets App to Top of Charts". GottaBeMobile.com. Archived from the original on 2 April 2016. Retrieved 10 May 2014. and third most popular for all iPhone apps at the start of the following month. In 2016, Outfit7 removed the chat feature along with the camera function from the app due to this controversy, though this decision was met with criticism.
Scale-space axioms
In image processing and computer vision, a scale space framework can be used to represent an image as a family of gradually smoothed images. This framework is very general and a variety of scale space representations exist. A typical approach for choosing a particular type of scale space representation is to establish a set of scale-space axioms, describing basic properties of the desired scale-space representation and often chosen so as to make the representation useful in practical applications. Once established, the axioms narrow the possible scale-space representations to a smaller class, typically with only a few free parameters. A set of standard scale space axioms, discussed below, leads to the linear Gaussian scale-space, which is the most common type of scale space used in image processing and computer vision. == Scale space axioms for the linear scale-space representation == The linear scale space representation L ( x , y , t ) = ( T t f ) ( x , y ) = g ( x , y , t ) ∗ f ( x , y ) {\displaystyle L(x,y,t)=(T_{t}f)(x,y)=g(x,y,t)f(x,y)} of signal f ( x , y ) {\displaystyle f(x,y)} obtained by smoothing with the Gaussian kernel g ( x , y , t ) {\displaystyle g(x,y,t)} satisfies a number of properties 'scale-space axioms' that make it a special form of multi-scale representation: linearity T t ( a f + b h ) = a T t f + b T t h {\displaystyle T_{t}(af+bh)=aT_{t}f+bT_{t}h} where f {\displaystyle f} and h {\displaystyle h} are signals while a {\displaystyle a} and b {\displaystyle b} are constants, shift invariance T t S ( Δ x , Δ y ) f = S ( Δ x , Δ y ) T t f {\displaystyle T_{t}S_{(\Delta x,\Delta _{y})}f=S_{(\Delta x,\Delta _{y})}T_{t}f} where S ( Δ x , Δ y ) {\displaystyle S_{(\Delta x,\Delta _{y})}} denotes the shift (translation) operator ( S ( Δ x , Δ y ) f ) ( x , y ) = f ( x − Δ x , y − Δ y ) {\displaystyle (S_{(\Delta x,\Delta _{y})}f)(x,y)=f(x-\Delta x,y-\Delta y)} semi-group structure g ( x , y , t 1 ) ∗ g ( x , y , t 2 ) = g ( x , y , t 1 + t 2 ) {\displaystyle g(x,y,t_{1})g(x,y,t_{2})=g(x,y,t_{1}+t_{2})} with the associated cascade smoothing property L ( x , y , t 2 ) = g ( x , y , t 2 − t 1 ) ∗ L ( x , y , t 1 ) {\displaystyle L(x,y,t_{2})=g(x,y,t_{2}-t_{1})L(x,y,t_{1})} existence of an infinitesimal generator A {\displaystyle A} ∂ t L ( x , y , t ) = ( A L ) ( x , y , t ) {\displaystyle \partial _{t}L(x,y,t)=(AL)(x,y,t)} non-creation of local extrema (zero-crossings) in one dimension, non-enhancement of local extrema in any number of dimensions ∂ t L ( x , y , t ) ≤ 0 {\displaystyle \partial _{t}L(x,y,t)\leq 0} at spatial maxima and ∂ t L ( x , y , t ) ≥ 0 {\displaystyle \partial _{t}L(x,y,t)\geq 0} at spatial minima, rotational symmetry g ( x , y , t ) = h ( x 2 + y 2 , t ) {\displaystyle g(x,y,t)=h(x^{2}+y^{2},t)} for some function h {\displaystyle h} , scale invariance g ^ ( ω x , ω y , t ) = h ^ ( ω x φ ( t ) , ω x φ ( t ) ) {\displaystyle {\hat {g}}(\omega _{x},\omega _{y},t)={\hat {h}}({\frac {\omega _{x}}{\varphi (t)}},{\frac {\omega _{x}}{\varphi (t)}})} for some functions φ {\displaystyle \varphi } and h ^ {\displaystyle {\hat {h}}} where g ^ {\displaystyle {\hat {g}}} denotes the Fourier transform of g {\displaystyle g} , positivity g ( x , y , t ) ≥ 0 {\displaystyle g(x,y,t)\geq 0} , normalization ∫ x = − ∞ ∞ ∫ y = − ∞ ∞ g ( x , y , t ) d x d y = 1 {\displaystyle \int _{x=-\infty }^{\infty }\int _{y=-\infty }^{\infty }g(x,y,t)\,dx\,dy=1} . In fact, it can be shown that the Gaussian kernel is a unique choice given several different combinations of subsets of these scale-space axioms: most of the axioms (linearity, shift-invariance, semigroup) correspond to scaling being a semigroup of shift-invariant linear operator, which is satisfied by a number of families integral transforms, while "non-creation of local extrema" for one-dimensional signals or "non-enhancement of local extrema" for higher-dimensional signals are the crucial axioms which relate scale-spaces to smoothing (formally, parabolic partial differential equations), and hence select for the Gaussian. The Gaussian kernel is also separable in Cartesian coordinates, i.e. g ( x , y , t ) = g ( x , t ) g ( y , t ) {\displaystyle g(x,y,t)=g(x,t)\,g(y,t)} . Separability is, however, not counted as a scale-space axiom, since it is a coordinate dependent property related to issues of implementation. In addition, the requirement of separability in combination with rotational symmetry per se fixates the smoothing kernel to be a Gaussian. There exists a generalization of the Gaussian scale-space theory to more general affine and spatio-temporal scale-spaces. In addition to variabilities over scale, which original scale-space theory was designed to handle, this generalized scale-space theory also comprises other types of variabilities, including image deformations caused by viewing variations, approximated by local affine transformations, and relative motions between objects in the world and the observer, approximated by local Galilean transformations. In this theory, rotational symmetry is not imposed as a necessary scale-space axiom and is instead replaced by requirements of affine and/or Galilean covariance. The generalized scale-space theory leads to predictions about receptive field profiles in good qualitative agreement with receptive field profiles measured by cell recordings in biological vision. In the computer vision, image processing and signal processing literature there are many other multi-scale approaches, using wavelets and a variety of other kernels, that do not exploit or require the same requirements as scale space descriptions do; please see the article on related multi-scale approaches. There has also been work on discrete scale-space concepts that carry the scale-space properties over to the discrete domain; see the article on scale space implementation for examples and references.
XLNet
The XLNet was an autoregressive Transformer designed as an improvement over BERT, with 340M parameters and trained on 33 billion words. It was released on 19 June 2019, under the Apache 2.0 license. It achieved state-of-the-art results on a variety of natural language processing tasks, including language modeling, question answering, and natural language inference. == Architecture == The main idea of XLNet is to model language autoregressively like the GPT models, but allow for all possible permutations of a sentence. Concretely, consider the following sentence:My dog is cute.In standard autoregressive language modeling, the model would be tasked with predicting the probability of each word, conditioned on the previous words as its context: We factorize the joint probability of a sequence of words x 1 , … , x T {\displaystyle x_{1},\ldots ,x_{T}} using the chain rule: Pr ( x 1 , … , x T ) = Pr ( x 1 ) Pr ( x 2 | x 1 ) Pr ( x 3 | x 1 , x 2 ) … Pr ( x T | x 1 , … , x T − 1 ) . {\displaystyle \Pr(x_{1},\ldots ,x_{T})=\Pr(x_{1})\Pr(x_{2}|x_{1})\Pr(x_{3}|x_{1},x_{2})\ldots \Pr(x_{T}|x_{1},\ldots ,x_{T-1}).} For example, the sentence "My dog is cute" is factorized as: Pr ( My , dog , is , cute ) = Pr ( My ) Pr ( dog | My ) Pr ( is | My , dog ) Pr ( cute | My , dog , is ) . {\displaystyle \Pr({\text{My}},{\text{dog}},{\text{is}},{\text{cute}})=\Pr({\text{My}})\Pr({\text{dog}}|{\text{My}})\Pr({\text{is}}|{\text{My}},{\text{dog}})\Pr({\text{cute}}|{\text{My}},{\text{dog}},{\text{is}}).} Schematically, we can write it as
EasyA
EasyA is a web3 technology company and education platform based in London (United Kingdom), founded in 2022 by Phil Kwok and Dom Kwok. EasyA was officially launched in 2022, focusing on web3 technologies. This community was influenced by the founders' experiences during the COVID-19 pandemic and early collaborations with universities and other educational institutions. Subsequently, the community was used as a foundation for developing Web3-related initiatives, including the organisation of EasyA's first Web3 hackathon in 2022. The EasyA app has over one million users and provides educational content on various blockchain technologies. EasyA Labs is a separate initiative focused on developing products intended to improve accessibility to cryptocurrency for a broader audience.
Pooling layer
In neural networks, a pooling layer is a kind of network layer that downsamples and aggregates information that is dispersed among many vectors into fewer vectors. It has several uses. It removes redundant information, thus reducing the amount of computation and memory required, which makes the model more robust to small variations in the input; and it increases the receptive field of neurons in later layers in the network. == Convolutional neural network pooling == Pooling is most commonly used in convolutional neural networks (CNN). Below is a description of pooling in 2-dimensional CNNs. The generalization to n-dimensions is immediate. As notation, we consider a tensor x ∈ R H × W × C {\displaystyle x\in \mathbb {R} ^{H\times W\times C}} , where H {\displaystyle H} is height, W {\displaystyle W} is width, and C {\displaystyle C} is the number of channels. A pooling layer outputs a tensor y ∈ R H ′ × W ′ × C ′ {\displaystyle y\in \mathbb {R} ^{H'\times W'\times C'}} . We define two variables f , s {\displaystyle f,s} called "filter size" (aka "kernel size") and "stride". Sometimes, it is necessary to use a different filter size and stride for horizontal and vertical directions. In such cases, we define 4 variables: f H , f W , s H , s W {\displaystyle f_{H},f_{W},s_{H},s_{W}} . The receptive field of an entry in the output tensor, y {\displaystyle y} , are all the entries in x {\displaystyle x} that can affect that entry. === Max pooling === Max Pooling (MaxPool) is commonly used in CNNs to reduce the spatial dimensions of feature maps. Define M a x P o o l ( x | f , s ) 0 , 0 , 0 = max ( x 0 : f − 1 , 0 : f − 1 , 0 ) {\displaystyle \mathrm {MaxPool} (x|f,s)_{0,0,0}=\max(x_{0:f-1,0:f-1,0})} where 0 : f − 1 {\displaystyle 0:f-1} means the range 0 , 1 , … , f − 1 {\displaystyle 0,1,\dots ,f-1} . Note that we need to avoid the off-by-one error. The next input is M a x P o o l ( x | f , s ) 1 , 0 , 0 = max ( x s : s + f − 1 , 0 : f − 1 , 0 ) {\displaystyle \mathrm {MaxPool} (x|f,s)_{1,0,0}=\max(x_{s:s+f-1,0:f-1,0})} and so on. The receptive field of y i , j , c {\displaystyle y_{i,j,c}} is x i s + f − 1 , j s + f − 1 , c {\displaystyle x_{is+f-1,js+f-1,c}} , so in general, M a x P o o l ( x | f , s ) i , j , c = m a x ( x i s : i s + f − 1 , j s : j s + f − 1 , c ) {\displaystyle \mathrm {MaxPool} (x|f,s)_{i,j,c}=\mathrm {max} (x_{is:is+f-1,js:js+f-1,c})} If the horizontal and vertical filter size and strides differ, then in general, M a x P o o l ( x | f , s ) i , j , c = m a x ( x i s H : i s H + f H − 1 , j s W : j s W + f W − 1 , c ) {\displaystyle \mathrm {MaxPool} (x|f,s)_{i,j,c}=\mathrm {max} (x_{is_{H}:is_{H}+f_{H}-1,js_{W}:js_{W}+f_{W}-1,c})} More succinctly, we can write y k = max ( { x k ′ | k ′ in the receptive field of k } ) {\displaystyle y_{k}=\max(\{x_{k'}|k'{\text{ in the receptive field of }}k\})} . If H {\displaystyle H} is not expressible as k s + f {\displaystyle ks+f} where k {\displaystyle k} is an integer, then for computing the entries of the output tensor on the boundaries, max pooling would attempt to take as inputs variables off the tensor. In this case, how those non-existent variables are handled depends on the padding conditions, illustrated on the right. Global Max Pooling (GMP) is a specific kind of max pooling where the output tensor has shape R C {\displaystyle \mathbb {R} ^{C}} and the receptive field of y c {\displaystyle y_{c}} is all of x 0 : H , 0 : W , c {\displaystyle x_{0:H,0:W,c}} . That is, it takes the maximum over each entire channel. It is often used just before the final fully connected layers in a CNN classification head. === Average pooling === Average pooling (AvgPool) is similarly defined A v g P o o l ( x | f , s ) i , j , c = a v e r a g e ( x i s : i s + f − 1 , j s : j s + f − 1 , c ) = 1 f 2 ∑ k ∈ i s : i s + f − 1 ∑ l ∈ j s : j s + f − 1 x k , l , c {\displaystyle \mathrm {AvgPool} (x|f,s)_{i,j,c}=\mathrm {average} (x_{is:is+f-1,js:js+f-1,c})={\frac {1}{f^{2}}}\sum _{k\in is:is+f-1}\sum _{l\in js:js+f-1}x_{k,l,c}} Global Average Pooling (GAP) is defined similarly to GMP. It was first proposed in Network-in-Network. Similarly to GMP, it is often used just before the final fully connected layers in a CNN classification head. === Interpolations === There are some interpolations of max pooling and average pooling. Mixed Pooling is a linear sum of max pooling and average pooling. That is, M i x e d P o o l ( x | f , s , w ) = w M a x P o o l ( x | f , s ) + ( 1 − w ) A v g P o o l ( x | f , s ) {\displaystyle \mathrm {MixedPool} (x|f,s,w)=w\mathrm {MaxPool} (x|f,s)+(1-w)\mathrm {AvgPool} (x|f,s)} where w ∈ [ 0 , 1 ] {\displaystyle w\in [0,1]} is either a hyperparameter, a learnable parameter, or randomly sampled anew every time. Lp Pooling is similar to average pooling, but uses Lp norm average instead of average: y k = ( 1 N ∑ k ′ in the receptive field of k | x k ′ | p ) 1 / p {\displaystyle y_{k}=\left({\frac {1}{N}}\sum _{k'{\text{ in the receptive field of }}k}|x_{k'}|^{p}\right)^{1/p}} where N {\displaystyle N} is the size of receptive field, and p ≥ 1 {\displaystyle p\geq 1} is a hyperparameter. If all activations are non-negative, then average pooling is the case of p = 1 {\displaystyle p=1} , and max pooling is the case of p → ∞ {\displaystyle p\to \infty } . Square-root pooling is the case of p = 2 {\displaystyle p=2} . Stochastic pooling samples a random activation x k ′ {\displaystyle x_{k'}} from the receptive field with probability x k ′ ∑ k ″ x k ″ {\displaystyle {\frac {x_{k'}}{\sum _{k''}x_{k''}}}} . It is the same as average pooling in expectation. Softmax pooling is like max pooling, but uses softmax, i.e. ∑ k ′ e β x k ′ x k ′ ∑ k ″ e β x k ″ {\displaystyle {\frac {\sum _{k'}e^{\beta x_{k'}}x_{k'}}{\sum _{k''}e^{\beta x_{k''}}}}} where β > 0 {\displaystyle \beta >0} . Average pooling is the case of β ↓ 0 {\displaystyle \beta \downarrow 0} , and max pooling is the case of β ↑ ∞ {\displaystyle \beta \uparrow \infty } Local Importance-based Pooling generalizes softmax pooling by ∑ k ′ e g ( x k ′ ) x k ′ ∑ k ″ e g ( x k ″ ) {\displaystyle {\frac {\sum _{k'}e^{g(x_{k'})}x_{k'}}{\sum _{k''}e^{g(x_{k''})}}}} where g {\displaystyle g} is a learnable function. === Other poolings === Spatial pyramidal pooling applies max pooling (or any other form of pooling) in a pyramid structure. That is, it applies global max pooling, then applies max pooling to the image divided into 4 equal parts, then 16, etc. The results are then concatenated. It is a hierarchical form of global pooling, and similar to global pooling, it is often used just before a classification head. Region of Interest Pooling (also known as RoI pooling) is a variant of max pooling used in R-CNNs for object detection. It is designed to take an arbitrarily-sized input matrix, and output a fixed-sized output matrix. Covariance pooling computes the covariance matrix of the vectors { x k , l , 0 : C − 1 } k ∈ i s : i s + f − 1 , l ∈ j s : j s + f − 1 {\displaystyle \{x_{k,l,0:C-1}\}_{k\in is:is+f-1,l\in js:js+f-1}} which is then flattened to a C 2 {\displaystyle C^{2}} -dimensional vector y i , j , 0 : C 2 − 1 {\displaystyle y_{i,j,0:C^{2}-1}} . Global covariance pooling is used similarly to global max pooling. As average pooling computes the average, which is a first-degree statistic, and covariance is a second-degree statistic, covariance pooling is also called "second-order pooling". It can be generalized to higher-order poolings. Blur Pooling means applying a blurring method before downsampling. For example, the Rect-2 blur pooling means taking an average pooling at f = 2 , s = 1 {\displaystyle f=2,s=1} , then taking every second pixel (identity with s = 2 {\displaystyle s=2} ). == Vision Transformer pooling == In Vision Transformers (ViT), there are the following common kinds of poolings. BERT-like pooling uses a dummy [CLS] token, "classification". For classification, the output at [CLS] is the classification token, which is then processed by a LayerNorm-feedforward-softmax module into a probability distribution, which is the network's prediction of class probability distribution. This is the one used by the original ViT and Masked Autoencoder. Global average pooling (GAP) does not use the dummy token, but simply takes the average of all output tokens as the classification token. It was mentioned in the original ViT as being equally good. Multihead attention pooling (MAP) applies a multi headed attention block to pooling. Specifically, it takes as input a list of vectors x 1 , x 2 , … , x n {\displaystyle x_{1},x_{2},\dots ,x_{n}} , which might be thought of as the output vectors of a layer of a ViT. It then applies a feedforward layer F F N {\displaystyle \mathrm {FFN} } on each vector, resulting in a matrix V = [ F F N ( v 1 ) , … , F F N ( v n ) ] {\displaystyle V=[\mathrm {FFN} (v_{1}),\dots ,\mathrm {FFN} (v_{n})]} . This is then sent to a multi-headed attention, resulting in M u l t i h e a d e d A