CatDV is a media asset manager program for handling multimedia production workflows developed by Square Box Systems. Quantum Corporation acquired Square Box Systems in 2020. == Versions == The full family of CatDV Products is as follows: CatDV Standalone Products CatDV Professional Edition CatDV Pegasus CatDV Networked Products CatDV Essential - entry level server product CatDV Enterprise Server - for MySQL databases and most common server platforms including Linux, Windows and Mac OS X CatDV Pegasus Server - adds features such as high performance full-text indexing, access control lists, and more CatDV Worker Node - automated workflow and transcoding engine CatDV Web Client - provides access to the CatDV database via a web browser. There is no need to install special software on the desktop, making it easy to deploy to a large number of users. CatDV Professional Edition & Pegasus Clients - designed to support the multi-user capabilities of the CatDV Enterprise and Workgroup Servers from the desktop Using plugins and scripting, which often require additional professional services support to set up, complex integrations with a wide variety of third party systems (including archive, cloud storage, and artificial intelligence) are possible. == Awards == CatDV won two awards in 2010, a blue ribbon from Creative COW Magazine and a "Best of Show Vidy Award" from Videography. In April 2012 Square Box won a Queen's Award for Enterprise for CatDV.
Foveated imaging
Foveated imaging is a digital image processing technique in which the image resolution, or amount of detail, varies across the image according to one or more "fixation points". A fixation point indicates the highest resolution region of the image and corresponds to the center of the eye's retina, the fovea. The location of a fixation point may be specified in many ways. For example, when viewing an image on a computer monitor, one may specify a fixation using a pointing device, like a computer mouse. Eye trackers which precisely measure the eye's position and movement are also commonly used to determine fixation points in perception experiments. When the display is manipulated with the use of an eye tracker, this is known as a gaze contingent display. Fixations may also be determined automatically using computer algorithms. Some common applications of foveated imaging include imaging sensor hardware and image compression. For descriptions of these and other applications, see the list below. Miniaturized foveated imaging systems can be realized by high-resolution 3D printing of multi-lens objectives directly on a CMOS (Complementary metal-oxide-semiconductor) chip. Foveated imaging is also commonly referred to as space variant imaging or gaze contingent imaging. == Applications == === Compression === Contrast sensitivity falls off dramatically as one moves from the center of the retina to the periphery. In lossy image compression, one may take advantage of this fact in order to compactly encode images. If one knows the viewer's approximate point of gaze, one may reduce the amount of information contained in the image as the distance from the point of gaze increases. Because the fall-off in the eye's resolution is dramatic, the potential reduction in display information can be substantial. Also, foveation encoding may be applied to the image before other types of image compression are applied and therefore can result in a multiplicative reduction. === Foveated sensors === Foveated sensors are multiresolution hardware devices that allow image data to be collected with higher resolution concentrated at a fixation point. An advantage to using foveated sensor hardware is that the image collection and encoding can occur much faster than in a system that post-processes a high resolution image in software. === Simulation === Foveated imaging has been used to simulate visual fields with arbitrary spatial resolution. For example, one may present video containing a blurred region representing a scotoma. By using an eye-tracker and holding the blurred region fixed relative to the viewer's gaze, the viewer will have a visual experience similar to that of a person with an actual scotoma. === Video gaming === Foveated rendering is a rendering optimization technique which uses an eye tracker integrated with a virtual reality headset to reduce the rendering workload by greatly reducing the image quality in the peripheral vision (outside of the zone gazed by the fovea).. However, other than the near-eye displays (e.g., virtual reality headset), foveated rendering is also suitable for large high-resolution display walls, desktop monitor, and even for smart phones. Over the time different foveated rendering techniques are proposed, for instance, adaptive resolution, geometric simplification, shading simplification and chromatic degradation, spatio-temporal deterioration . If we consider the variable sample distribution of physically-based rendering under the shader (e.g., hit/miss etc.), then this degradation strategies are applied on overall foveated rendering. At the CES 2016, SensoMotoric Instruments (SMI) demoed a new 250 Hz eye tracking system and a working foveated rendering solution. It resulted from a partnership with camera sensor manufacturer Omnivision who provided the camera hardware for the new system. The Apple Vision Pro mixed reality headset features dynamic foveated rendering provided by its visionOS operating system. === Quality assessment === Foveated imaging may be useful in providing a subjective image quality measure. Traditional image quality measures, such as peak signal-to-noise ratio, are typically performed on fixed resolution images and do not take into account some aspects of the human visual system, like the change in spatial resolution across the retina. A foveated quality index may therefore more accurately determine image quality as perceived by humans. === Image database retrieval === In databases that contain very high resolution images, such as a satellite image database, it may be desirable to interactively retrieve images in order to reduce retrieval time. Foveated imaging allows one to scan low resolution images and retrieve only high resolution portions as they are needed. This is sometimes called progressive transmission. == Example images ==
Sudip Roy (computer scientist)
Sudip Roy is a computer scientist and technology executive. He is the co-founder and chief technology officer of Adaption. He has worked on large-scale machine learning systems at organizations including Google DeepMind and Cohere. == Education == Roy earned a PhD in Computer Science from Cornell University. He holds a B.Tech in Computer Science and Engineering from the Indian Institute of Technology (IIT), Kharagpur. == Career == Sudip worked at Google Brain (now part of Google DeepMind) on systems research and large-scale data management. During his tenure, he contributed to infrastructure projects including Pathways and TensorFlow Extended, which support training and inference workflows for production machine learning models. He later served as Senior Director of Engineering at Cohere, leading work on inference infrastructure and fine-tuning systems. In late 2025, he co-founded the company Adaption Labs with Sara Hooker. The company focuses on developing AI systems designed for continuous learning and adaptation. Roy’s research spans systems for AI and AI for systems, including work on optimizing system performance and compilers. His publications have appeared in conferences such as MLSys, NeurIPS, SIGMOD, and KDD. He has been a program committee member or reviewer for the conferences SIGMOD, VLDB, ICDE, and MLSys. == Awards == He is the recipient of the MLSys Outstanding Paper Award (2022) and the SIGMOD Best Paper Award (2011). He holds multiple patents in machine learning systems, including methods for learned graph optimizations and neural network-based device placement.
Trigram tagger
In computational linguistics, a trigram tagger is a statistical method for automatically identifying words as being nouns, verbs, adjectives, adverbs, etc. based on second order Markov models that consider triples of consecutive words. It is trained on a text corpus as a method to predict the next word, taking the product of the probabilities of unigram, bigram and trigram. In speech recognition, algorithms utilizing trigram-tagger score better than those algorithms utilizing IIMM tagger but less well than Net tagger. The description of the trigram tagger is provided by Brants (2000).
Restricted Boltzmann machine
A restricted Boltzmann machine (RBM) (also called a restricted Sherrington–Kirkpatrick model with external field or restricted stochastic Ising–Lenz–Little model) is a generative stochastic artificial neural network that can learn a probability distribution over its set of inputs. RBMs were initially proposed under the name Harmonium by Paul Smolensky in 1986, and rose to prominence after Geoffrey Hinton and collaborators used fast learning algorithms for them in the mid-2000s. RBMs have found applications in dimensionality reduction, classification, collaborative filtering, feature learning, topic modelling, immunology, and even many‑body quantum mechanics. They can be trained in either supervised or unsupervised ways, depending on the task. As their name implies, RBMs are a variant of Boltzmann machines, with the restriction that their neurons must form a bipartite graph: a pair of nodes from each of the two groups of units (commonly referred to as the "visible" and "hidden" units respectively) may have a symmetric connection between them; and there are no connections between nodes within a group. By contrast, "unrestricted" Boltzmann machines may have connections between hidden units. This restriction allows for more efficient training algorithms than are available for the general class of Boltzmann machines, in particular the gradient-based contrastive divergence algorithm. Restricted Boltzmann machines can also be used in deep learning networks. In particular, deep belief networks can be formed by "stacking" RBMs and optionally fine-tuning the resulting deep network with gradient descent and backpropagation. == Structure == The standard type of RBM has binary-valued (Boolean) hidden and visible units, and consists of a matrix of weights W {\displaystyle W} of size m × n {\displaystyle m\times n} . Each weight element ( w i , j ) {\displaystyle (w_{i,j})} of the matrix is associated with the connection between the visible (input) unit v i {\displaystyle v_{i}} and the hidden unit h j {\displaystyle h_{j}} . In addition, there are bias weights (offsets) a i {\displaystyle a_{i}} for v i {\displaystyle v_{i}} and b j {\displaystyle b_{j}} for h j {\displaystyle h_{j}} . Given the weights and biases, the energy of a configuration (pair of Boolean vectors) (v,h) is defined as E ( v , h ) = − ∑ i a i v i − ∑ j b j h j − ∑ i ∑ j v i w i , j h j {\displaystyle E(v,h)=-\sum _{i}a_{i}v_{i}-\sum _{j}b_{j}h_{j}-\sum _{i}\sum _{j}v_{i}w_{i,j}h_{j}} or, in matrix notation, E ( v , h ) = − a T v − b T h − v T W h . {\displaystyle E(v,h)=-a^{\mathrm {T} }v-b^{\mathrm {T} }h-v^{\mathrm {T} }Wh.} This energy function is analogous to that of a Hopfield network. As with general Boltzmann machines, the joint probability distribution for the visible and hidden vectors is defined in terms of the energy function as follows, P ( v , h ) = 1 Z e − E ( v , h ) {\displaystyle P(v,h)={\frac {1}{Z}}e^{-E(v,h)}} where Z {\displaystyle Z} is a partition function defined as the sum of e − E ( v , h ) {\displaystyle e^{-E(v,h)}} over all possible configurations, which can be interpreted as a normalizing constant to ensure that the probabilities sum to 1. The marginal probability of a visible vector is the sum of P ( v , h ) {\displaystyle P(v,h)} over all possible hidden layer configurations, P ( v ) = 1 Z ∑ { h } e − E ( v , h ) {\displaystyle P(v)={\frac {1}{Z}}\sum _{\{h\}}e^{-E(v,h)}} , and vice versa. Since the underlying graph structure of the RBM is bipartite (meaning there are no intra-layer connections), the hidden unit activations are mutually independent given the visible unit activations. Conversely, the visible unit activations are mutually independent given the hidden unit activations. That is, for m visible units and n hidden units, the conditional probability of a configuration of the visible units v, given a configuration of the hidden units h, is P ( v | h ) = ∏ i = 1 m P ( v i | h ) {\displaystyle P(v|h)=\prod _{i=1}^{m}P(v_{i}|h)} . Conversely, the conditional probability of h given v is P ( h | v ) = ∏ j = 1 n P ( h j | v ) {\displaystyle P(h|v)=\prod _{j=1}^{n}P(h_{j}|v)} . The individual activation probabilities are given by P ( h j = 1 | v ) = σ ( b j + ∑ i = 1 m w i , j v i ) {\displaystyle P(h_{j}=1|v)=\sigma \left(b_{j}+\sum _{i=1}^{m}w_{i,j}v_{i}\right)} and P ( v i = 1 | h ) = σ ( a i + ∑ j = 1 n w i , j h j ) {\displaystyle \,P(v_{i}=1|h)=\sigma \left(a_{i}+\sum _{j=1}^{n}w_{i,j}h_{j}\right)} where σ {\displaystyle \sigma } denotes the logistic sigmoid. The visible units of Restricted Boltzmann Machine can be multinomial, although the hidden units are Bernoulli. In this case, the logistic function for visible units is replaced by the softmax function P ( v i k = 1 | h ) = exp ( a i k + Σ j W i j k h j ) Σ k ′ = 1 K exp ( a i k ′ + Σ j W i j k ′ h j ) {\displaystyle P(v_{i}^{k}=1|h)={\frac {\exp(a_{i}^{k}+\Sigma _{j}W_{ij}^{k}h_{j})}{\Sigma _{k'=1}^{K}\exp(a_{i}^{k'}+\Sigma _{j}W_{ij}^{k'}h_{j})}}} where K is the number of discrete values that the visible values have. They are applied in topic modeling, and recommender systems. === Relation to other models === Restricted Boltzmann machines are a special case of Boltzmann machines and Markov random fields. The graphical model of RBMs corresponds to that of factor analysis. == Training algorithm == Restricted Boltzmann machines are trained to maximize the product of probabilities assigned to some training set V {\displaystyle V} (a matrix, each row of which is treated as a visible vector v {\displaystyle v} ), arg max W ∏ v ∈ V P ( v ) {\displaystyle \arg \max _{W}\prod _{v\in V}P(v)} or equivalently, to maximize the expected log probability of a training sample v {\displaystyle v} selected randomly from V {\displaystyle V} : arg max W E [ log P ( v ) ] {\displaystyle \arg \max _{W}\mathbb {E} \left[\log P(v)\right]} The algorithm most often used to train RBMs, that is, to optimize the weight matrix W {\displaystyle W} , is the contrastive divergence (CD) algorithm due to Hinton, originally developed to train PoE (product of experts) models. The algorithm performs Gibbs sampling and is used inside a gradient descent procedure (similar to the way backpropagation is used inside such a procedure when training feedforward neural nets) to compute weight update. The basic, single-step contrastive divergence (CD-1) procedure for a single sample can be summarized as follows: Take a training sample v, compute the probabilities of the hidden units and sample a hidden activation vector h from this probability distribution. Compute the outer product of v and h and call this the positive gradient. From h, sample a reconstruction v' of the visible units, then resample the hidden activations h' from this. (Gibbs sampling step) Compute the outer product of v' and h' and call this the negative gradient. Let the update to the weight matrix W {\displaystyle W} be the positive gradient minus the negative gradient, times some learning rate: Δ W = ϵ ( v h T − v ′ h ′ T ) {\displaystyle \Delta W=\epsilon (vh^{\mathsf {T}}-v'h'^{\mathsf {T}})} . Update the biases a and b analogously: Δ a = ϵ ( v − v ′ ) {\displaystyle \Delta a=\epsilon (v-v')} , Δ b = ϵ ( h − h ′ ) {\displaystyle \Delta b=\epsilon (h-h')} . A Practical Guide to Training RBMs written by Hinton can be found on his homepage. == Stacked Restricted Boltzmann Machine == The difference between the Stacked Restricted Boltzmann Machines and RBM is that RBM has lateral connections within a layer that are prohibited to make analysis tractable. On the other hand, the Stacked Boltzmann consists of a combination of an unsupervised three-layer network with symmetric weights and a supervised fine-tuned top layer for recognizing three classes. The usage of Stacked Boltzmann is to understand Natural languages, retrieve documents, image generation, and classification. These functions are trained with unsupervised pre-training and/or supervised fine-tuning. Unlike the undirected symmetric top layer, with a two-way unsymmetric layer for connection for RBM. The restricted Boltzmann's connection is three-layers with asymmetric weights, and two networks are combined into one. Stacked Boltzmann does share similarities with RBM, the neuron for Stacked Boltzmann is a stochastic binary Hopfield neuron, which is the same as the Restricted Boltzmann Machine. The energy from both Restricted Boltzmann and RBM is given by Gibb's probability measure: E = − 1 2 ∑ i , j w i j s i s j + ∑ i θ i s i {\displaystyle E=-{\frac {1}{2}}\sum _{i,j}{w_{ij}{s_{i}}{s_{j}}}+\sum _{i}{\theta _{i}}{s_{i}}} . The training process of Restricted Boltzmann is similar to RBM. Restricted Boltzmann train one layer at a time and approximate equilibrium state with a 3-segment pass, not performing back propagation. Restricted Boltzmann uses both supervised and unsupervised on different RBM for pre-training for classification and recognition. The training uses contrastive divergence with
Abiquo Enterprise Edition
Abiquo Hybrid Cloud Management Platform is a web-based cloud computing software platform developed by Abiquo. Written entirely in Java, it is used to build, integrate and manage public and private clouds in homogeneous environments. Users can deploy and manage servers, storage system and network and virtual devices. It also supports LDAP integration. == Hypervisors == Abiquo supports five hypervisor systems. VMware ESXi Microsoft Hyper-V Citrix XenServer Oracle VM Server for x86 KVM From version 3.1, it also supports multiple public cloud providers: Amazon AWS Rackspace Google Compute Engine HP Cloud ElasticHosts DigitalOcean Abiquo version 3.2 added: Microsoft Azure Abiquo version 3.4 added: Support for Docker hosts, adding multi-tenant networking, storage management and private registry management for Docker SoftLayer CloudSigma Later versions continued to add features including autoscaling on any cloud, integration to VMware NSX and OpenStack Neutron for software defined networking, guest config with cloud-init and integrated monitoring driving guest automation. == Storage services == Abiquo supports any vendor for hypervisor storage, and also supports tiered storage pools, enabling storage-as-a-service from specific vendors and technologies including: NFS Generic iSCSI NetApp Nexenta == SAAS version == In April 2014 Abiquo launched Abiquo anyCloud, a SAAS version of the Abiquo Hybrid Cloud Platform software. This version lets users manage public cloud resources from: Amazon AWS Microsoft Azure IBM SoftLayer DigitalOcean Rackspace Open Cloud (an OpenStack cloud) HP Public Cloud (an OpenStack cloud) Google Compute Engine ElasticHosts Additional security and process features include workflow, to have an enterprise administrator electronically sign off on changes, an audit trail of activity and the ability to share cloud accounts among and enterprise team in a secure way. == Reviews and awards == Finalist for the 2015 Cloud Awards Finalist for the 2015 UK Cloud Awards in the category Cloud Management Product of the Year EMA Radar for Private Cloud platforms 2013 Global Telecoms Business Innovation Summit and Awards 2013 (with Interoute) EuroCloud UK Awards
Additive smoothing
In statistics, additive smoothing, also called Laplace smoothing or Lidstone smoothing, is a technique used to smooth count data, eliminating issues caused by certain values having 0 occurrences. Given a set of observation counts x = ⟨ x 1 , x 2 , … , x d ⟩ {\displaystyle \mathbf {x} =\langle x_{1},x_{2},\ldots ,x_{d}\rangle } from a d {\displaystyle d} -dimensional multinomial distribution with N {\displaystyle N} trials, a "smoothed" version of the counts gives the estimator θ ^ i = x i + α N + α d ( i = 1 , … , d ) , {\displaystyle {\hat {\theta }}_{i}={\frac {x_{i}+\alpha }{N+\alpha d}}\qquad (i=1,\ldots ,d),} where the smoothed count x ^ i = N θ ^ i {\displaystyle {\hat {x}}_{i}=N{\hat {\theta }}_{i}} , and the "pseudocount" α > 0 is a smoothing parameter, with α = 0 corresponding to no smoothing (this parameter is explained in § Pseudocount below). Additive smoothing is a type of shrinkage estimator, as the resulting estimate will be between the empirical probability (relative frequency) x i / N {\displaystyle x_{i}/N} and the uniform probability 1 / d . {\displaystyle 1/d.} Common choices for α are 0 (no smoothing), +1⁄2 (the Jeffreys prior), or 1 (Laplace's rule of succession), but the parameter may also be set empirically based on the observed data. From a Bayesian point of view, this corresponds to the expected value of the posterior distribution, using a symmetric Dirichlet distribution with parameter α as a prior distribution. In the special case where the number of categories is 2, this is equivalent to using a beta distribution as the conjugate prior for the parameters of the binomial distribution. == History == Laplace came up with this smoothing technique when he tried to estimate the chance that the sun will rise tomorrow. His rationale was that even given a large sample of days with the rising sun, we still can not be completely sure that the sun will still rise tomorrow (known as the sunrise problem). == Pseudocount == A pseudocount is an amount (not generally an integer, despite its name) added to the number of observed cases in order to change the expected probability in a model of those data, when not known to be zero. It is so named because, roughly speaking, a pseudo-count of value α {\displaystyle \alpha } weighs into the posterior distribution similarly to each category having an additional count of α {\displaystyle \alpha } . If the number of occurrences of each item i {\displaystyle i} is x i {\displaystyle x_{i}} out of N {\displaystyle N} samples, the empirical probability of event i {\displaystyle i} is p i , empirical = x i N , {\displaystyle p_{i,{\text{empirical}}}={\frac {x_{i}}{N}},} but the posterior probability when additively smoothed is p i , α -smoothed = x i + α N + α d , {\displaystyle p_{i,\alpha {\text{-smoothed}}}={\frac {x_{i}+\alpha }{N+\alpha d}},} as if to increase each count x i {\displaystyle x_{i}} by α {\displaystyle \alpha } a priori. Depending on the prior knowledge, which is sometimes a subjective value, a pseudocount may have any non-negative finite value. It may only be zero (or the possibility ignored) if impossible by definition, such as the possibility of a decimal digit of π being a letter, or a physical possibility that would be rejected and so not counted, such as a computer printing a letter when a valid program for π is run, or excluded and not counted because of no interest, such as if only interested in the zeros and ones. Generally, there is also a possibility that no value may be computable or observable in a finite time (see the halting problem). But at least one possibility must have a non-zero pseudocount, otherwise no prediction could be computed before the first observation. The relative values of pseudocounts represent the relative prior expected probabilities of their possibilities. The sum of the pseudocounts, which may be very large, represents the estimated weight of the prior knowledge compared with all the actual observations (one for each) when determining the expected probability. In any observed data set or sample there is the possibility, especially with low-probability events and with small data sets, of a possible event not occurring. Its observed frequency is therefore zero, apparently implying a probability of zero. This oversimplification is inaccurate and often unhelpful, particularly in probability-based machine learning techniques such as artificial neural networks and hidden Markov models. By artificially adjusting the probability of rare (but not impossible) events so those probabilities are not exactly zero, zero-frequency problems are avoided. Also see Cromwell's rule. === Choice of pseudocount === ==== Weakly informative prior ==== One common approach is to add 1 to each observed number of events, including the zero-count possibilities. This is sometimes called Laplace's rule of succession. This approach is equivalent to assuming a uniform prior distribution over the probabilities for each possible event (spanning the simplex where each probability is between 0 and 1, and they all sum to 1). Using the Jeffreys prior approach, a pseudocount of one half should be added to each possible outcome. Pseudocounts should be set to one or one-half only when there is no prior knowledge at all – see the principle of indifference. However, given appropriate prior knowledge, the sum should be adjusted in proportion to the expectation that the prior probabilities should be considered correct, despite evidence to the contrary – see further analysis. Higher values are appropriate inasmuch as there is prior knowledge of the true values (for a mint-condition coin, say); lower values inasmuch as there is prior knowledge that there is probable bias, but of unknown degree (for a bent coin, say). ==== Frequentist interval ==== One way to motivate pseudocounts, particularly for binomial data, is via a formula for the midpoint of an interval estimate, particularly a binomial proportion confidence interval. The best-known is due to Edwin Bidwell Wilson, in Wilson (1927): the midpoint of the Wilson score interval corresponding to z {\displaystyle z} standard deviations on either side is n S + z n + 2 z {\displaystyle {\frac {n_{S}+z}{n+2z}}} Taking z = 2 {\displaystyle z=2} standard deviations to approximate a 95% confidence interval ( z ≈ 1.96 {\displaystyle z\approx 1.96} ) yields pseudocount of 2 for each outcome, so 4 in total, colloquially known as the "plus four rule": n S + 2 n + 4 {\displaystyle {\frac {n_{S}+2}{n+4}}} This is also the midpoint of the Agresti–Coull interval (Agresti & Coull 1998). ==== Known incidence rates ==== Often the bias of an unknown trial population is tested against a control population with known parameters (incidence rates) μ = ⟨ μ 1 , μ 2 , … , μ d ⟩ . {\displaystyle {\boldsymbol {\mu }}=\langle \mu _{1},\mu _{2},\ldots ,\mu _{d}\rangle .} In this case the uniform probability 1 / d {\displaystyle 1/d} should be replaced by the known incidence rate of the control population μ i {\displaystyle \mu _{i}} to calculate the smoothed estimator: θ ^ i = x i + μ i α d N + α d ( i = 1 , … , d ) . {\displaystyle {\hat {\theta }}_{i}={\frac {x_{i}+\mu _{i}\alpha d}{N+\alpha d}}\qquad (i=1,\ldots ,d).} As a consistency check, if the empirical estimator happens to equal the incidence rate, i.e. μ i = x i / N , {\displaystyle \mu _{i}=x_{i}/N,} the smoothed estimator is independent of α {\displaystyle \alpha } and also equals the incidence rate. == Applications == === Classification === Additive smoothing is commonly a component of naive Bayes classifiers. === Statistical language modelling === In a bag of words model of natural language processing and information retrieval, the data consists of the number of occurrences of each word in a document. Additive smoothing allows the assignment of non-zero probabilities to words which do not occur in the sample. Studies have shown that additive smoothing is more effective than other probability smoothing methods in several retrieval tasks such as language-model-based pseudo-relevance feedback and recommender systems.