Product-family engineering (PFE), also known as product-line engineering (PLE), is based on the ideas of "domain engineering" created by the Software Engineering Institute, a term coined by James Neighbors in his 1980 dissertation at University of California, Irvine. Software product lines are quite common in our daily lives, but before a product family can be successfully established, an extensive process has to be followed. This process is known as product-family engineering. Product-family engineering can be defined as a method that creates an underlying architecture of an organization's product platform. It provides an architecture that is based on commonality as well as planned variabilities. The various product variants can be derived from the basic product family, which creates the opportunity to reuse and differentiate on products in the family. Product-family engineering is conceptually similar to the widespread use of vehicle platforms in the automotive industry. Product-family engineering is a relatively new approach to the creation of new products, recently evolving to Model-Based Product Line Engineering (MBPLE), emphasizing the centrality of a model-centric approach in PLE. It focuses on the process of engineering new products in such a way that it is possible to reuse product components and apply variability with decreased costs and time. Product-family engineering is all about reusing components and structures as much as possible, according to the ISO/IEC 26550/2015 and the latest ISO/IEC 26580/2021 that introduced the concept of feature-based Product Line Engineering. Several studies have proven that using a product-family engineering approach for product development can have several benefits. Here is a list of some of them: Higher productivity Higher quality Faster time-to-market Lower labor needs The Nokia case mentioned below also illustrates these benefits. In 2025 the publishing of the book Model-Based Product Line Engineering (MBPLE): The feature-based path to product lines success by Marco Forlingieri, Tim Weilkiens and Hugo Guillermo Chalé-Gongora formalized the foundation of the discipline, including best practices and new industrial cases. == Overall process == The product family engineering process consists of several phases. The three main phases are: Phase 1: Product management Phase 2: Domain engineering Phase 3: Product engineering The process has been modeled on a higher abstraction level. This has the advantage that it can be applied to all kinds of product lines and families, not only software. The model can be applied to any product family. Figure 1 (below) shows a model of the entire process. Below, the process is described in detail. The process description contains elaborations of the activities and the important concepts being used. All concepts printed in italic are explained in Table 1. === Phase 1: product management === The first phase is the starting up of the whole process. In this phase some important aspects are defined especially with regard to economic aspects. This phase is responsible for outlining market strategies and defining a scope, which tells what should and should not be inside the product family. ==== Evaluate business visioning ==== During this first activity all context information relevant for defining the scope of the product line is collected and evaluated. It is important to define a clear market strategy and take external market information into account, such as consumer demands. The activity should deliver a context document that contains guidelines, constraints and the product strategy. ==== Define product line scope ==== Scoping techniques are applied to define which aspects are within the scope. This is based upon the previous step in the process, where external factors have been taken into account. The output is a product portfolio description, which includes a list of current and future products and also a product roadmap. It can be argued whether phase 1, product management, is part of the product-family-engineering process, because it could be seen as an individual business process that is more focused on the management aspects instead of the product aspect. However phase 2 needs some important input from this phase, as a large piece of the scope is defined in this phase. So from this point of view it is important to include the product-management phase (phase 1) into the entire process as a base for the domain-engineering process. === Phase 2: domain engineering === During the domain-engineering phases, the variable and common requirements are gathered for the whole product line. The goal is to establish a reusable platform. The output of this phase is a set of common and variable requirements for all products in the product line. ==== Analyze domain requirements ==== This activity includes all activities for analyzing the domain with regard to concept requirements. The requirements are categorized and split up into two new activities. The output is a document with the domain analysis. As can be seen in Figure 1 the process of defining common requirements is a parallel process with defining variable requirements. Both activities take place at the same time. ==== Define common requirements ==== Includes all activities for eliciting and documenting the common requirements of the product line, resulting in a document with reusable common requirements. ==== Define variable requirements ==== Includes all activities for eliciting and documenting the variable requirements of the product line, resulting in a document with variable requirements. ==== Design domain ==== This process step consists of activities for defining the reference architecture of the product line. This generates an abstract structure for all products in the product line. ==== Implement domain ==== During this step a detailed design of the reusable components and the implementation of these components are created. ==== Test domain ==== Validates and verifies the reusability of components. Components are tested against their specifications. After successful testing of all components in different use cases and scenarios, the domain engineering phase has been completed. === Phase 3: product engineering === In the final phase a product X is being engineered. This product X uses the commonalities and variability from the domain engineering phase, so product X is being derived from the platform established in the domain engineering phase. It basically takes all common requirements and similarities from the preceding phase plus its own variable requirements. Using the base from the domain engineering phase and the individual requirements of the product engineering phase a complete and new product can be built. After the product has been fully tested and approved, the product X can be delivered. ==== Define product requirements ==== Developing the product requirements specification for the individual product and reuse the requirements from the preceding phase. ==== Design product ==== All activities for producing the product architecture. Makes use of the reference architecture from the step "design domain", it selects and configures the required parts of the reference architecture and incorporates product specific adaptations. ==== Build product ==== During this process the product is built, using selections and configurations of the reusable components. ==== Test product ==== During this step the product is verified and validated against its specifications. A test report gives information about all tests that were carried out, this gives an overview of possible errors in the product. If the product in the next step is not accepted, the process will loop back to "build product", in Figure 1 this is indicated as "[unsatisfied]". ==== Deliver and support product ==== The final step is the acceptance of the final product. If it has been successfully tested and approved to be complete, it can be delivered. If the product does not satisfy to the specifications, it has to be rebuilt and tested again. The next figure shows the overall process of product-family engineering as described above. It is a full process overview with all concepts attached to the different steps. == Process data diagram == On the left side the entire process from the top to bottom has been drawn. All activities on the left side are linked to the concepts on the right side through dotted lines. Every concept has a number, which reflects the association with other concepts. == List of concepts == Below the list with concepts will be explained. Most concept definitions are extracted from Pohl, Bockle, & Linden (2005) and also some new definitions have been added. Table 1: List of concepts == Example == There are some good examples of the use of product family engineering, which were quite successful. The abstract model of product family engineering allows different kinds of uses, most of them are related to the consumer electronics m
Semantic neural network
Semantic neural network (SNN) is based on John von Neumann's neural network [von Neumann, 1966] and Nikolai Amosov M-Network. There are limitations to a link topology for the von Neumann’s network but SNN accept a case without these limitations. Only logical values can be processed, but SNN accept that fuzzy values can be processed too. All neurons into the von Neumann network are synchronized by tacts. For further use of self-synchronizing circuit technique SNN accepts neurons can be self-running or synchronized. In contrast to the von Neumann network there are no limitations for topology of neurons for semantic networks. It leads to the impossibility of relative addressing of neurons as it was done by von Neumann. In this case an absolute readdressing should be used. Every neuron should have a unique identifier that would provide a direct access to another neuron. Of course, neurons interacting by axons-dendrites should have each other's identifiers. An absolute readdressing can be modulated by using neuron specificity as it was realized for biological neural networks. There’s no description for self-reflectiveness and self-modification abilities into the initial description of semantic networks [Dudar Z.V., Shuklin D.E., 2000]. But in [Shuklin D.E. 2004] a conclusion had been drawn about the necessity of introspection and self-modification abilities in the system. For maintenance of these abilities a concept of pointer to neuron is provided. Pointers represent virtual connections between neurons. In this model, bodies and signals transferring through the neurons connections represent a physical body, and virtual connections between neurons are representing an astral body. It is proposed to create models of artificial neuron networks on the basis of virtual machine supporting the opportunity for paranormal effects. SNN is generally used for natural language processing. == Related models == Computational creativity Semantic hashing Semantic Pointer Architecture Sparse distributed memory
Online machine learning
In computer science, online machine learning is a method of machine learning in which data becomes available in a sequential order and is used to update the best predictor for future data at each step, as opposed to batch learning techniques which generate the best predictor by learning on the entire training data set at once. Online learning is a common technique used in areas of machine learning where it is computationally infeasible to train over the entire dataset, requiring the need of out-of-core algorithms. It is also used in situations where it is necessary for the algorithm to dynamically adapt to new patterns in the data, or when the data itself is generated as a function of time, e.g., prediction of prices in the financial international markets. Online learning algorithms may be prone to catastrophic interference, a problem that can be addressed by incremental learning approaches. Online machine learning algorithms find applications in a wide variety of fields such as sponsored search to maximize ad revenue, portfolio optimization, shortest path prediction (with stochastic weights, e.g. traffic on roads for a maps application), spam filtering, real-time fraud detection, dynamic pricing for e-commerce, etc. There is also growing interest in usage of online learning paradigms for LLMs to enable continuous, real-time adaptation after the initial training. == Introduction == In the setting of supervised learning, a function of f : X → Y {\displaystyle f:X\to Y} is to be learned, where X {\displaystyle X} is thought of as a space of inputs and Y {\displaystyle Y} as a space of outputs, that predicts well on instances that are drawn from a joint probability distribution p ( x , y ) {\displaystyle p(x,y)} on X × Y {\displaystyle X\times Y} . In reality, the learner never knows the true distribution p ( x , y ) {\displaystyle p(x,y)} over instances. Instead, the learner usually has access to a training set of examples ( x 1 , y 1 ) , … , ( x n , y n ) {\displaystyle (x_{1},y_{1}),\ldots ,(x_{n},y_{n})} . In this setting, the loss function is given as V : Y × Y → R {\displaystyle V:Y\times Y\to \mathbb {R} } , such that V ( f ( x ) , y ) {\displaystyle V(f(x),y)} measures the difference between the predicted value f ( x ) {\displaystyle f(x)} and the true value y {\displaystyle y} . The ideal goal is to select a function f ∈ H {\displaystyle f\in {\mathcal {H}}} , where H {\displaystyle {\mathcal {H}}} is a space of functions called a hypothesis space, so that some notion of total loss is minimized. Depending on the type of model (statistical or adversarial), one can devise different notions of loss, which lead to different learning algorithms. == Statistical view of online learning == In statistical learning models, the training sample ( x i , y i ) {\displaystyle (x_{i},y_{i})} are assumed to have been drawn from the true distribution p ( x , y ) {\displaystyle p(x,y)} and the objective is to minimize the expected "risk" I [ f ] = E [ V ( f ( x ) , y ) ] = ∫ V ( f ( x ) , y ) d p ( x , y ) . {\displaystyle I[f]=\mathbb {E} [V(f(x),y)]=\int V(f(x),y)\,dp(x,y)\ .} A common paradigm in this situation is to estimate a function f ^ {\displaystyle {\hat {f}}} through empirical risk minimization or regularized empirical risk minimization (usually Tikhonov regularization). The choice of loss function here gives rise to several well-known learning algorithms such as regularized least squares and support vector machines. A purely online model in this category would learn based on just the new input ( x t + 1 , y t + 1 ) {\displaystyle (x_{t+1},y_{t+1})} , the current best predictor f t {\displaystyle f_{t}} and some extra stored information (which is usually expected to have storage requirements independent of training data size). For many formulations, for example nonlinear kernel methods, true online learning is not possible, though a form of hybrid online learning with recursive algorithms can be used where f t + 1 {\displaystyle f_{t+1}} is permitted to depend on f t {\displaystyle f_{t}} and all previous data points ( x 1 , y 1 ) , … , ( x t , y t ) {\displaystyle (x_{1},y_{1}),\ldots ,(x_{t},y_{t})} . In this case, the space requirements are no longer guaranteed to be constant since it requires storing all previous data points, but the solution may take less time to compute with the addition of a new data point, as compared to batch learning techniques. A common strategy to overcome the above issues is to learn using mini-batches, which process a small batch of b ≥ 1 {\displaystyle b\geq 1} data points at a time, this can be considered as pseudo-online learning for b {\displaystyle b} much smaller than the total number of training points. Mini-batch techniques are used with repeated passing over the training data to obtain optimized out-of-core versions of machine learning algorithms, for example, stochastic gradient descent. When combined with backpropagation, this is currently the de facto training method for training artificial neural networks. === Example: linear least squares === The simple example of linear least squares is used to explain a variety of ideas in online learning. The ideas are general enough to be applied to other settings, for example, with other convex loss functions. === Batch learning === Consider the setting of supervised learning with f {\displaystyle f} being a linear function to be learned: f ( x j ) = ⟨ w , x j ⟩ = w ⋅ x j {\displaystyle f(x_{j})=\langle w,x_{j}\rangle =w\cdot x_{j}} where x j ∈ R d {\displaystyle x_{j}\in \mathbb {R} ^{d}} is a vector of inputs (data points) and w ∈ R d {\displaystyle w\in \mathbb {R} ^{d}} is a linear filter vector. The goal is to compute the filter vector w {\displaystyle w} . To this end, a square loss function V ( f ( x j ) , y j ) = ( f ( x j ) − y j ) 2 = ( ⟨ w , x j ⟩ − y j ) 2 {\displaystyle V(f(x_{j}),y_{j})=(f(x_{j})-y_{j})^{2}=(\langle w,x_{j}\rangle -y_{j})^{2}} is used to compute the vector w {\displaystyle w} that minimizes the empirical loss I n [ w ] = ∑ j = 1 n V ( ⟨ w , x j ⟩ , y j ) = ∑ j = 1 n ( x j T w − y j ) 2 {\displaystyle I_{n}[w]=\sum _{j=1}^{n}V(\langle w,x_{j}\rangle ,y_{j})=\sum _{j=1}^{n}(x_{j}^{\mathsf {T}}w-y_{j})^{2}} where y j ∈ R . {\displaystyle y_{j}\in \mathbb {R} .} Let X {\displaystyle X} be the i × d {\displaystyle i\times d} data matrix and y ∈ R i {\displaystyle y\in \mathbb {R} ^{i}} is the column vector of target values after the arrival of the first i {\displaystyle i} data points. Assuming that the covariance matrix Σ i = X T X {\displaystyle \Sigma _{i}=X^{\mathsf {T}}X} is invertible (otherwise it is preferential to proceed in a similar fashion with Tikhonov regularization), the best solution f ∗ ( x ) = ⟨ w ∗ , x ⟩ {\displaystyle f^{}(x)=\langle w^{},x\rangle } to the linear least squares problem is given by w ∗ = ( X T X ) − 1 X T y = Σ i − 1 ∑ j = 1 i x j y j . {\displaystyle w^{}=(X^{\mathsf {T}}X)^{-1}X^{\mathsf {T}}y=\Sigma _{i}^{-1}\sum _{j=1}^{i}x_{j}y_{j}.} Now, calculating the covariance matrix Σ i = ∑ j = 1 i x j x j T {\displaystyle \Sigma _{i}=\sum _{j=1}^{i}x_{j}x_{j}^{\mathsf {T}}} takes time O ( i d 2 ) {\displaystyle O(id^{2})} , inverting the d × d {\displaystyle d\times d} matrix takes time O ( d 3 ) {\displaystyle O(d^{3})} , while the rest of the multiplication takes time O ( d 2 ) {\displaystyle O(d^{2})} , giving a total time of O ( i d 2 + d 3 ) {\displaystyle O(id^{2}+d^{3})} . When there are n {\displaystyle n} total points in the dataset, to recompute the solution after the arrival of every datapoint i = 1 , … , n {\displaystyle i=1,\ldots ,n} , the naive approach will have a total complexity O ( n 2 d 2 + n d 3 ) {\displaystyle O(n^{2}d^{2}+nd^{3})} . Note that when storing the matrix Σ i {\displaystyle \Sigma _{i}} , then updating it at each step needs only adding x i + 1 x i + 1 T {\displaystyle x_{i+1}x_{i+1}^{\mathsf {T}}} , which takes O ( d 2 ) {\displaystyle O(d^{2})} time, reducing the total time to O ( n d 2 + n d 3 ) = O ( n d 3 ) {\displaystyle O(nd^{2}+nd^{3})=O(nd^{3})} , but with an additional storage space of O ( d 2 ) {\displaystyle O(d^{2})} to store Σ i {\displaystyle \Sigma _{i}} . === Online learning: recursive least squares === The recursive least squares (RLS) algorithm considers an online approach to the least squares problem. It can be shown that by initialising w 0 = 0 ∈ R d {\displaystyle \textstyle w_{0}=0\in \mathbb {R} ^{d}} and Γ 0 = I ∈ R d × d {\displaystyle \textstyle \Gamma _{0}=I\in \mathbb {R} ^{d\times d}} , the solution of the linear least squares problem given in the previous section can be computed by the following iteration: Γ i = Γ i − 1 − Γ i − 1 x i x i T Γ i − 1 1 + x i T Γ i − 1 x i {\displaystyle \Gamma _{i}=\Gamma _{i-1}-{\frac {\Gamma _{i-1}x_{i}x_{i}^{\mathsf {T}}\Gamma _{i-1}}{1+x_{i}^{\mathsf {T}}\Gamma _{i-1}x_{i}}}} w i = w i − 1 − Γ i x i ( x i T w i − 1 − y i ) {\displaystyle w_{i}=w_{i-1}-\Gamma _{i}x_{i}\left(x_{i}^{\mathsf {T}}w_{
Cross-entropy
In information theory, the cross-entropy between two probability distributions p {\displaystyle p} and q {\displaystyle q} , over the same underlying set of events, measures the average number of bits needed to identify an event drawn from the set when the coding scheme used for the set is optimized for an estimated probability distribution q {\displaystyle q} , rather than the true distribution p {\displaystyle p} . == Definition == The cross-entropy of the distribution q {\displaystyle q} relative to a distribution p {\displaystyle p} over a given set is defined as follows: H ( p , q ) = − E p [ log q ] , {\displaystyle H(p,q)=-\operatorname {E} _{p}[\log q],} where E p [ ⋅ ] {\displaystyle \operatorname {E} _{p}[\cdot ]} is the expected value operator with respect to the distribution p {\displaystyle p} . The definition may be formulated using the Kullback–Leibler divergence D K L ( p ∥ q ) {\displaystyle D_{\mathrm {KL} }(p\parallel q)} , divergence of p {\displaystyle p} from q {\displaystyle q} (also known as the relative entropy of p {\displaystyle p} with respect to q {\displaystyle q} ). H ( p , q ) = H ( p ) + D K L ( p ∥ q ) , {\displaystyle H(p,q)=H(p)+D_{\mathrm {KL} }(p\parallel q),} where H ( p ) {\displaystyle H(p)} is the entropy of p {\displaystyle p} . For discrete probability distributions p {\displaystyle p} and q {\displaystyle q} with the same support X {\displaystyle {\mathcal {X}}} , this means The situation for continuous distributions is analogous. We have to assume that p {\displaystyle p} and q {\displaystyle q} are absolutely continuous with respect to some reference measure r {\displaystyle r} (usually r {\displaystyle r} is a Lebesgue measure on a Borel σ-algebra). Let P {\displaystyle P} and Q {\displaystyle Q} be probability density functions of p {\displaystyle p} and q {\displaystyle q} with respect to r {\displaystyle r} . Then − ∫ X P ( x ) log Q ( x ) d x = E p [ − log Q ] , {\displaystyle -\int _{\mathcal {X}}P(x)\,\log Q(x)\,\mathrm {d} x=\operatorname {E} _{p}[-\log Q],} and therefore NB: The notation H ( p , q ) {\displaystyle H(p,q)} is also used for a different concept, the joint entropy of p {\displaystyle p} and q {\displaystyle q} . == Motivation == In information theory, the Kraft–McMillan theorem establishes that any directly decodable coding scheme for coding a message to identify one value x i {\displaystyle x_{i}} out of a set of possibilities { x 1 , … , x n } {\displaystyle \{x_{1},\ldots ,x_{n}\}} can be seen as representing an implicit probability distribution q ( x i ) = ( 1 2 ) ℓ i {\displaystyle q(x_{i})=\left({\frac {1}{2}}\right)^{\ell _{i}}} over { x 1 , … , x n } {\displaystyle \{x_{1},\ldots ,x_{n}\}} , where ℓ i {\displaystyle \ell _{i}} is the length of the code for x i {\displaystyle x_{i}} in bits. Therefore, cross-entropy can be interpreted as the expected message-length per datum when a wrong distribution q {\displaystyle q} is assumed while the data actually follows a distribution p {\displaystyle p} . That is why the expectation is taken over the true probability distribution p {\displaystyle p} and not q . {\displaystyle q.} Indeed the expected message-length under the true distribution p {\displaystyle p} is E p [ ℓ ] = − E p [ ln q ( x ) ln ( 2 ) ] = − E p [ log 2 q ( x ) ] = − ∑ x i p ( x i ) log 2 q ( x i ) = − ∑ x p ( x ) log 2 q ( x ) = H ( p , q ) . {\displaystyle {\begin{aligned}\operatorname {E} _{p}[\ell ]&=-\operatorname {E} _{p}\left[{\frac {\ln {q(x)}}{\ln(2)}}\right]\\[1ex]&=-\operatorname {E} _{p}\left[\log _{2}{q(x)}\right]\\[1ex]&=-\sum _{x_{i}}p(x_{i})\,\log _{2}q(x_{i})\\[1ex]&=-\sum _{x}p(x)\,\log _{2}q(x)=H(p,q).\end{aligned}}} == Estimation == There are many situations where cross-entropy needs to be measured but the distribution of p {\displaystyle p} is unknown. An example is language modeling, where a model is created based on a training set T {\displaystyle T} , and then its cross-entropy is measured on a test set to assess how accurate the model is in predicting the test data. In this example, p {\displaystyle p} is the true distribution of words in any corpus, and q {\displaystyle q} is the distribution of words as predicted by the model. Since the true distribution is unknown, cross-entropy cannot be directly calculated. In these cases, an estimate of cross-entropy is calculated using the following formula: H ( T , q ) = − ∑ i = 1 N 1 N log 2 q ( x i ) {\displaystyle H(T,q)=-\sum _{i=1}^{N}{\frac {1}{N}}\log _{2}q(x_{i})} where N {\displaystyle N} is the size of the test set, and q ( x ) {\displaystyle q(x)} is the probability of event x {\displaystyle x} estimated from the training set. In other words, q ( x i ) {\displaystyle q(x_{i})} is the probability estimate of the model that the i-th word of the text is x i {\displaystyle x_{i}} . The sum is averaged over the N {\displaystyle N} words of the test. This is a Monte Carlo estimate of the true cross-entropy, where the test set is treated as samples from p ( x ) {\displaystyle p(x)} . == Relation to maximum likelihood == The cross entropy arises in classification problems when introducing a logarithm in the guise of the log-likelihood function. This section concerns the estimation of the probabilities of different discrete outcomes. To this end, denote a parametrized family of distributions by q θ {\displaystyle q_{\theta }} , with θ {\displaystyle \theta } subject to the optimization effort. Consider a given finite sequence of N {\displaystyle N} values x i {\displaystyle x_{i}} from a training set, obtained from conditionally independent sampling. The likelihood assigned to any considered parameter θ {\displaystyle \theta } of the model is then given by the product over all probabilities q θ ( X = x i ) {\displaystyle q_{\theta }(X=x_{i})} . Repeated occurrences are possible, leading to equal factors in the product. If the count of occurrences of the value equal to x {\displaystyle x} is denoted by # x {\displaystyle \#x} , then the frequency of that value equals # x / N {\displaystyle \#x/N} . If p ( X = x ) {\displaystyle p(X=x)} is the underlying probability distribution, for large N {\displaystyle N} we expect p ( X = x ) ≈ # x / N {\displaystyle p(X=x)\approx \#x/N} , by the law of large numbers. Writing our likelihood function as the product of observations from the distribution q θ {\displaystyle q_{\theta }} : L ( θ ; x ) = ∏ i q θ ( X = x i ) = ∏ x q θ ( X = x ) # x ≈ ∏ x q θ ( X = x ) N ⋅ p ( X = x ) = exp log [ ∏ x q θ ( X = x ) N ⋅ p ( X = x ) ] = exp ( ∑ x N ⋅ p ( X = x ) log q θ ( X = x ) ) , {\displaystyle {\begin{aligned}{\mathcal {L}}(\theta ;{\mathbf {x} })&=\prod _{i}q_{\theta }(X=x_{i})=\prod _{x}q_{\theta }(X=x)^{\#x}\\&\approx \prod _{x}q_{\theta }(X=x)^{N\cdot p(X=x)}=\exp \log \left[\prod _{x}q_{\theta }(X=x)^{N\cdot p(X=x)}\right]\\&=\exp \left(\sum _{x}N\cdot p(X=x)\log q_{\theta }(X=x)^{}\right),\end{aligned}}} where we have used the calculation rules for the logarithm in the final line. Notice how the exponent contains a − H ( p , q θ ) {\displaystyle -H(p,q_{\theta })} term. Taking the logarithm of both sides gives: log L ( θ ; x ) = − N ⋅ H ( p , q θ ) . {\displaystyle \log {\mathcal {L}}(\theta ;{\mathbf {x} })=-N\cdot H(p,q_{\theta }).} Since the logarithm is a monotonically increasing function, the maximizing value of θ {\displaystyle \theta } is unaffected by this final step. Similarly, the maximizing value of θ {\displaystyle \theta } is unaffected by the factor of N {\displaystyle N} . So we observe that the likelihood maximization amounts to minimization of the cross-entropy. == Cross-entropy minimization == Cross-entropy minimization is frequently used in optimization and rare-event probability estimation. When comparing a distribution q {\displaystyle q} against a fixed reference distribution p {\displaystyle p} , cross-entropy and KL divergence are identical up to an additive constant (since p {\displaystyle p} is fixed): According to the Gibbs' inequality, both take on their minimal values when p = q {\displaystyle p=q} , which is 0 {\displaystyle 0} for KL divergence, and H ( p ) {\displaystyle \mathrm {H} (p)} for cross-entropy. In the engineering literature, the principle of minimizing KL divergence (Kullback's "Principle of Minimum Discrimination Information") is often called the Principle of Minimum Cross-Entropy (MCE), or Minxent. However, as discussed in the article Kullback–Leibler divergence, sometimes the distribution q {\displaystyle q} is the fixed prior reference distribution, and the distribution p {\displaystyle p} is optimized to be as close to q {\displaystyle q} as possible, subject to some constraint. In this case the two minimizations are not equivalent. This has led to some ambiguity in the literature, with some authors attempting to resolve the inconsistency by restating cross-entropy to be D K L ( p ∥ q ) {\displaystyle D_{\mathrm {KL} }(p\parallel q)} , rather than H (
Jubatus
Jubatus is an open-source online machine learning and distributed computing framework developed at Nippon Telegraph and Telephone and Preferred Infrastructure. Its features include classification, recommendation, regression, anomaly detection and graph mining. It supports many client languages, including C++, Java, Ruby and Python. It uses Iterative Parameter Mixture for distributed machine learning. == Notable Features == Jubatus supports: Multi-classification algorithms: Perceptron Passive Aggressive Confidence Weighted Adaptive Regularization of Weight Vectors Normal Herd Recommendation algorithms using: Inverted index Minhash Locality-sensitive hashing Regression algorithms: Passive Aggressive feature extraction method for natural language: n-gram Text segmentation
Web Dynpro
Web Dynpro (WD) is a web application technology developed by SAP SE that focuses on the development of server-side business applications. For modern releases (for instance as of NetWeaver 750, software layer SAP_UI) the user interface is rendered according to the HTML5 web standard. Since Netweaver 754 (software layer SAP_UI, ABAP Platform 1909) a touch enabled user interface is available. The newly released versions usually follow the SAP Fiori design principles. One of its main design features is that the user interface is defined in an entirely declarative manner. Web Dynpro applications can be developed using either the Java (Web Dynpro for Java, WDJ or WD4J) or ABAP (Web Dynpro ABAP, WDA or WD4A) development infrastructure. == Overview == The earliest version of Web Dynpro appeared in 2003 and was based on Java. This variant was released approximately 18 months before the ABAP variant. As of 2010, the Java variant of Web Dynpro was put into maintenance mode. WD follows a design architecture based on an interpretation of the MVC design pattern and uses a model driven development approach ("minimize coding, maximize design"). The Web Dynpro Framework is a server-side runtime environment into which many dedicated "hook methods" are available. The developer then places their own custom coding within these hook methods in order to implement the desired business functionality. These hook methods belong to one of the broad categories of either "life-cycle" and "round-trip"; that is, those methods that are concerned with the life-cycle of a software component (i.e. processing that takes place at start up and shut down etc.), and those methods that are concerned with processing the fixed sequence of events that take place during a client-initiated round trip to the server. Web Dynpro is aimed at the development of business applications that follow standardized UI principles, applications that connect to backend systems and which are scalable. Key Capabilities Declarative way of development: Web Dynpro offers a graphical and declarative means of UI development. UI controls, building blocks, views and windows are modeled, and the business logic can be coded separately. Separation of user interface and business logic: One advantage of Web Dynpro over SAP GUI is the separation between business logic and user interface, and the structured development process with less implementation effort. Support of stateful application: The state of the application is kept in the back-end. This leads to a reduced data transfer from ABAP server to browser and vice versa. Regarding Web Dynpro ABAP there is only one programming language (ABAP) and only one system necessary. Therefore, development can be easier and cost efficient.
Teaching dimension
In computational learning theory, the teaching dimension of a concept class C is defined to be max c ∈ C { w C ( c ) } {\displaystyle \max _{c\in C}\{w_{C}(c)\}} , where w C ( c ) {\displaystyle {w_{C}(c)}} is the minimum size of a witness set for c in C. Intuitively, this measures the number of instances that are needed to identify a concept in the class, using supervised learning with examples provided by a helpful teacher who is trying to convey the concept as succinctly as possible. This definition was formulated in 1995 by Sally Goldman and Michael Kearns, based on earlier work by Goldman, Ron Rivest, and Robert Schapire. The teaching dimension of a finite concept class can be used to give a lower and an upper bound on the membership query cost of the concept class. In Stasys Jukna's book "Extremal Combinatorics", a lower bound is given for the teaching dimension in general: Let C be a concept class over a finite domain X. If the size of C is greater than 2 k ( | X | k ) , {\displaystyle 2^{k}{|X| \choose k},} then the teaching dimension of C is greater than k. However, there are more specific teaching models that make assumptions about teacher or learner, and can get lower values for the teaching dimension. For instance, several models are the classical teaching (CT) model, the optimal teacher (OT) model, recursive teaching (RT), preference-based teaching (PBT), and non-clashing teaching (NCT).