Search references for LOGP MACHINE. Phrases containing LOGP MACHINE
See searches and references containing LOGP MACHINE!LOGP MACHINE
Topics referred to by the same term
LogP may refer to: the logarithm of the partition coefficient, in the physical sciences LogP machine, a model for parallel computation p-adic logarithm
LogP
The LogP machine is a model for parallel computation. It aims at being more practical than the PRAM model while still allowing for easy analysis of computation
LogP_machine
Average uncertainty in variable's states
is distributed according to p:š³ā[0,1], the entropy is H(X):=āāxāš³p(x)logā”p(x), where Ī£ denotes the sum over the variable's possible values. The choice
Entropy_(information_theory)
Class of artificial neural network
probability of a training sample v selected randomly from V: argā”maxWš¼[logā”P(v)] The algorithm most often used to train RBMs, that is, to optimize the
Restricted_Boltzmann_machine
Mathematical statistics distance measure
probability distribution P. Mathematically, it is defined as DKL(Pā„Q)=āxāš³P(x)logā”P(x)Q(x). A simple interpretation of the KL divergence of P from Q is the
KullbackāLeibler_divergence
Iterative method for finding maximum likelihood estimates in statistical models
estimates of the parameters ā š½(t)ā : Q(š½ā£š½(t))=Ešā¼p(ā |š,š½(t))[logā”p(š,š|š½)]:=ā«logā”p(š,š|š½)p(š|š,š½(t))dš Maximization step (M step): Find the
Expectationāmaximization algorithm
Expectationāmaximization_algorithm
Machine learning paradigm
distribution P(y|x) and the loss function is the negative log likelihood: L(y,Å·)=ālogā”P(y|x), then empirical risk minimization is equivalent to maximum likelihood
Supervised_learning
Pressure exerted by a vapor in thermodynamic equilibrium
basic form of the equation is: logā”P=AāBC+T and it can be transformed into this temperature-explicit form: T=BAālogā”PāC where: P is the absolute vapor
Vapor_pressure
Abstraction of parallel computer architecture
Computation), https://cs.brown.edu/~jes/book/ Archived 2016-11-05 at the Wayback Machine "1.3 A Parallel Programming Model". www.mcs.anl.gov. Retrieved 2024-03-21
Parallel_programming_model
Model for designing parallel algorithms
computing Concurrency (computer science) Dataflow programming Grid computing LogP machine Parallel computing Parallel programming model Leslie G. Valiant, A bridging
Bulk_synchronous_parallel
Airport in Austria
Pinkafeld Airport (German: Flugplatz Pinkafeld, ICAO: LOGP) is a private use airport located 2 kilometres (1Ā mi) north-northwest of Pinkafeld, Burgenland
Pinkafeld_Airport
Paradigm in machine learning
based on fit to both the labeled and unlabeled data, weighted by Ī»: argmaxĪ(logā”p({xi,yi}i=1l|Īø)+Ī»logā”p({xi}i=l+1l+u|Īø)) Another major class of methods attempts
Weak_supervision
Measure of lipophilicity and hydrophilicity
a common logarithm, i.e. ā logā”Pā (also log Pow or, less frequently, Log pOW): logā”P=logā”coSicwSi=logā”coSiālogā”cwSi ā logā”Pā is positive for lipophilic
Octanol-water partition coefficient
Octanol-water_partition_coefficient
Hypothesis in neuroscience
μ)} F(s,μ)āfreeāenergy=Eq[ālogā”p(s,Ļ|m)]āenergyāH[qμ(Ļ)]āentropy=ālogā”p(s|m)āsurprise+DKL[qμ(Ļ)||p(Ļ|s,m)]ādivergenceā„ālogā”p(s|m)āsurprise This induces
Free_energy_principle
Measure of dependence between two variables
yāš“p(X,Y)(x,y)logā”p(X,Y)(x,y)pX(x)pY(y)=āxāš³,yāš“p(X,Y)(x,y)logā”p(X,Y)(x,y)pX(x)āāxāš³,yāš“p(X,Y)(x,y)logā”pY(y)=āxāš³,yāš“pX(x)pYā£X=x(y)logā”pYā£X=x(y)āāxāš³
Mutual_information
Optimization and sampling technique
distribution p(Īøā£X)āp(Īø)āi=1Np(xiā£Īø) by updating the chain: ĪĪøt=εt2(ālogā”p(Īøt)+āi=1Nālogā”p(xtiā£Īøt))+Ī·t Stochastic gradient Langevin dynamics uses a modified
Stochastic gradient Langevin dynamics
Stochastic_gradient_Langevin_dynamics
Probabilistic classification algorithm
classifier when expressed in log-space: logā”p(Ckā£š±)ālogā”(p(Ck)āi=1npkixi)=logā”p(Ck)+āi=1nxiā logā”pki=b+š°kā¤š± where b=logā”p(Ck) and wki=logā”pki. Estimating the
Naive_Bayes_classifier
Mathematical methods used in Bayesian inference and machine learning
logā”P(š,š)]+logā”P(š) which can be rearranged to become logā”P(š)=DKL(Qā„P)āš¼š[logā”Q(š)ālogā”P(š,š)]=DKL(Qā„P)+ā(Q) As the log-evidence logā”P(š)
Variational_Bayesian_methods
Short chains of 2ā50 amino acids
molecular properties used for mapping include molecular weight, lipophilicity (logP and logD), topological polar surface area (TPSA), and hydrogen-bond dynamics
Peptide
Rule of thumb to predict if a chemical compound is likely to be an orally active drug
PMIDĀ 14554012. Free online calculations of Hydrogen bond donor/acceptor, mass and logP using ChemAxon's Marvin and Calculator Plugins ā requires Java Calculation
Lipinski's_rule_of_five
Scientific study of digital information
will be the entropy of the position of the piece. H(X,Y)=š¼X,Y[ālogā”p(x,y)]=āāx,yp(x,y)logā”p(x,y) Despite similar notation, joint entropy should not be confused
Information_theory
American computer scientist
Parallel Complexity within the LogP Model was based on her work with Karp and others introducing and analyzing the LogP machine, an abstract model of parallel
Eunice_Santos
Chemical compound
The lipophilicity of fluorenol (LogP 2.4) is higher than that of drugs like modafinil (LogP 1.7) and amphetamine (LogP 1.8), suggesting that it may penetrate
Fluorenol
Statistical model used in machine learning
the target distribution. The log likelihood of zK is (see derivation): logā”pK(zK)=logā”p0(z0)āāi=1Klogā”|detā”dfi(ziā1)dziā1| Learning probability distributions
Flow-based_generative_model
Numerical method
function can be simplified as below: ā(Ļ,μ,Ī£)=āi=1mlogā”p(x(i)ā£z(i);μ,Ī£)+logā”p(z(i);Ļ) Now the likelihood function can be maximized by making partial derivative
EM_algorithm_and_GMM_model
Boolean satisfiability is NP-complete and therefore that NP-complete problems exist
variables. There are O(p(n)2) Boolean variables, each encodable in space O(logā”p(n)). The number of clauses is O(p(n)3) so the size of B is O(logā”(p(n))p(n)3)
CookāLevin_theorem
Estimate of the importance of a word in a document
document frequency, P(t|D)=|{dāD:tād}|N, so that we can define idf as idf=ālogā”P(t|D)=logā”1P(t|D)=logā”N|{dāD:tād}| Namely, the inverse document frequency
Tfāidf
Principle in Bayesian statistics
reference measure or prior density m(x), typically by maximizing āā«p(x)logā”p(x)m(x)dx subject to the given constraints. This formulation is invariant
Principle_of_maximum_entropy
Probability distribution
H(X)=āāk=0āP(X=k)lnā”P(X=k)=āāk=0ā(1āp)kplnā”((1āp)kp)=āāk=0ā(1āp)kp[klnā”(1āp)+lnā”p]=ālogā”pā1āpplogā”(1āp) The entropy increases as the probability p decreases, reflecting
Geometric_distribution
Statistical model
given set of hyperparameters Īø. As such the log marginal likelihood is: logā”p(f(xā²)ā£Īø,x)=ā12(fš³(x)Kā1(Īø,x,xā²)f(xā²)+logā”detā”(K(Īø,x,xā²))+nlogā”2Ļ) and maximizing
Gaussian_process
Statistical distance measure
āMlogā”M+12[āPlogā”P+āQlogā”Q]=āāP2logā”MāāQ2logā”M+12[āPlogā”P+āQlogā”Q]=12āP(logā”Pālogā”M)+12āQ(logā”Qālogā”M)=JSD(Pā„Q) It follows from the above result that
JensenāShannon_divergence
Mathematical model used for classification or regression
could also be replaced by the log-loss equation below: llog(xi,yi,c(xi;w))=ālogā”p(yi|xi;w)=logā”Z(xi;w)āwTĻ(xi,yi) Since the log-loss is differentiable, a
Discriminative_model
Notion in statistics
divergence, between two distributions p and q can be written as KL(p:q)=ā«p(x)logā”p(x)q(x)dx. Now, consider a family of probability distributions f(x;Īø) parametrized
Fisher_information
Measure of difference between two points
KullbackāLeibler divergence DF(p,q)=āip(i)logā”p(i)q(i)āāip(i)+āiq(i) is generated by the negative entropy function F(p)=āip(i)logā”p(i) When restricted to the simplex
Bregman_divergence
Concept in drug design
alcohol), so the logarithm of the octanol-water partition coefficient, known as LogP, is used to predict the solubility of a potential oral drug. This coefficient
Druglikeness
Characterization of how many integers are prime
Ļ(x)=ākā„1āpĀ is primepkā¤x,logā”p. This is sometimes written as Ļ(x)=ānā¤xĪ(n), where Ī(n) is the von Mangoldt function, namely Ī(n)={logā”pĀ ifĀ n=pkĀ for some primeĀ pĀ and
Prime_number_theorem
Predictive chemical model
methods (known as "CLogP" and other variations). It has been shown that the logP of compound can be determined by the sum of its fragments; fragment-based
Quantitative structureāactivity relationship
Quantitative_structureāactivity_relationship
Model of interest rate curves
under risk-neutral pricing rules. Let's define the following process: Ytālogā”P(t,s)=āā«tsf(t,u)du The dynamics of Yt can be obtained through Leibniz's rule:
HeathāJarrowāMorton framework
HeathāJarrowāMorton_framework
Hungarian cheminformatics software company
been referenced in cheminformatics studies for predicting pKa values and logP values. Cheminformatics Computational chemistry Molecular modeling Drug discovery
Chemaxon
Algorithm for statistical inference on graphical models
E(š)=ālogā”āfjfj(xj). The free energy of the system is then F=UāH=āšP(š)E(š)+āšP(š)logā”P(š). It can then be shown that the points of convergence of the sum-product
Belief_propagation
High-performance forward error correction codes
yk=y1k) and to DEC2 (when yk=y2k). DEC1 yields a soft decision; i.e.: Ī(dk)=logā”p(dk=1)p(dk=0) and delivers it to DEC2. Ī(dk) is called the logarithm of the
Turbo_code
prefix sum algorithm to calculate the prefix sum with the optimal O(NPB+logā”P) I/O complexity. This algorithm simulates an optimal PRAM prefix sum algorithm
Parallel_external_memory
Measure of distance to normality
entropy of Y, with pY as its probability density function: h(Y)=āā«pY(u)logā”pY(u)du Negentropy is used in statistics and signal processing. It is related
Negentropy
Property of a thermodynamic system
in terms of a discrete set of probabilities pi so that:H(X)=āāi=1np(xi)logā”p(xi)where the base of the logarithm determines the units (for example, the
Entropy
Software application for cheminformatics
descriptors, charge descriptors, and more. Molecular properties: including LogP (octanol-water partition coefficient), molar refractivity, polar surface
AlvaDesc
Approximate interference technique in Bayesian networks
to be as large as possible; because it's a lower bound, getting closer logā”P improves the approximation of the log likelihood. By substituting in the
Variational_message_passing
Algorithmic optimization method
possible to simulate the results of P numerical comparisons using only O(logā”P) calls to the decision algorithm. Thus, the total time for parametric search
Parametric_search
Probability distribution
The conditional information entropy of Z, given X is S(šæ)=H(š|šæ)=Eš[ālogā”P(š|šæ)]=āi=1KāXilogā”Xi This function of X is a scalar random variable. If
Dirichlet_distribution
Two dwarf galaxies orbiting the Milky Way
home.insightbb.com. Archived from the original on 2003-04-29. "The (MK, logP) and (Mbol, log P) relations". aa.springer.de. Archived from the original
Magellanic_Clouds
Medication, stimulant and decongestant
and Leo54 found that blood-brain barrier penetration is optimal when the LogP values are in the range of 1.5-2.7, with a mean value of 2.1. An analysis
Ephedrine
Part of the history of physics
section, Shannon introduces an H function of the following form: H=āKāi=1kp(i)logā”p(i), where K is a positive constant. Shannon then states that "any quantity
History_of_entropy
Process forming a path from many random steps
ā(Xn=k)=2ān(nn+k2) for nā”k(mod2), and using Stirling's formula one has logā”ā(Xn=k)=ā12logā”(Ļn2)ān+k2logā”(1+kn)ānāk2logā”(1ākn)+o(1). Fixing the scaling
Random_walk
Quantity in information theory
Shannon entropy of the random variable X is defined as:H(X)Ā =Ā āxāpX(x)Ā logā”pX(x)=Ā āxpX(x)Ā IX(x)Ā Ā =defĀ Ā Eā”[IX(X)],by definition equal to the expected
Information_content
original URL status unknown (link) pdf Archived 16 March 2016 at the Wayback Machine Mandelbrot, B. B. (1983). The Fractal Geometry of Nature, p.48. New York:
List of fractals by Hausdorff dimension
List_of_fractals_by_Hausdorff_dimension
Algorithm that estimates unknowns from a series of measurements over time
implementation it is usually desirable to compute the log marginal likelihood ā=logā”p(š³) instead. Adopting the convention ā(ā1)=0, this can be done via the recursive
Kalman_filter
Grammar model in linguistics
calculates γ(i,j,v) to find the most probable parse tree ĻĢ and yields logā”P(x,ĻĢ|Īø). Memory and time complexity for general PCFG algorithms in RNA structure
Probabilistic context-free grammar
Probabilistic_context-free_grammar
Paradox arising from the question of what constitutes evidence for a statement
a black raven exceeds that provided by a non-black non-raven by about ālogā”P(Ba|RaHā¾). This is equal to the amount of additional information (in bits
Raven_paradox
Antibiotic medication
uncharged state. It is moderately water-soluble (approximately 3Ā mg/mL), with a logP of 0.55. The oxazolidinone pharmacophoreāthe chemical "template" essential
Linezolid
Evolutionary algorithm
maximizes a log-likelihood, such that mk+1=argā”maxmāi=1μwilogā”pš©(xi:Ī»ā£m) where logā”pš©(x)=ā12logā”detā”(2ĻC)ā12(xām)TCā1(xām) denotes the log-likelihood of x from
CMA-ES
Method of estimating the parameters of a statistical model, given observations
a uniform prior distribution āā”(Īø). In many practical applications in machine learning, maximum-likelihood estimation is used as the model for parameter
Maximum_likelihood_estimation
Algorithm for integer factorization
one random number to be drawn.) Stages. Compute a point Q:=āpā¤B1pālogā”B1/logā”pāP on the E. The product means that one loops over every prime pā¤B1; it plays
Lenstra elliptic-curve factorization
Lenstra_elliptic-curve_factorization
examples of compounds which are tropanes for prospective use in research Article on cocaine analogue research Archived 2006-09-22 at the Wayback Machine
List_of_cocaine_analogues
images Ļā(ā ), written as densities the log-posterior takes the form logā”p(Ļā Iā£ID)ālogā”p(IDā£Ļā I)+logā”ĻDiffV(Ļ)Ā . Maximum a posteriori estimation (MAP) estimation
Bayesian estimation of templates in computational anatomy
Bayesian_estimation_of_templates_in_computational_anatomy
travel, tourism, insurance
LOGP MACHINE
LOGP MACHINE
LOGP MACHINE
LOGP MACHINE
LOGP MACHINE
LOGP MACHINE
LOGP MACHINE
LOGP MACHINE
LOGP MACHINE
travel, tourism, insurance