Search
Search Funnelback University
- Refined by:
- Date: 2023
201 -
250 of
259
search results for katalk:za33 24 |u:mlg.eng.cam.ac.uk
where 0
match all words and 259
match some words.
Results that match 1 of 2 words
-
chu05a.dvi
https://mlg.eng.cam.ac.uk/zoubin/papers/chu05a.pdf27 Jan 2023: 25.732.24% 23.781.85% 23.751.74% 0.25960.0230 0.24110.0189 0.24110.0186Boston 25.561.98% 24.882.02% 24.491.85% 0.26720.0190 -
Data-Efficient Reinforcement Learning inContinuous State-Action…
https://mlg.eng.cam.ac.uk/pub/pdf/McaRas17.pdf13 Feb 2023: Bt|t N(Mt|t, V̄t|t), and Mt|t N(µmt|t, Σmt|t), (24). or equivalently the joint[Bt|tMt|t. ] -
coverage.eps
https://mlg.eng.cam.ac.uk/pub/pdf/SilHelGhaetal10.pdf13 Feb 2023: student course f aculty projectcornell 0.87 0.82 0.87 0.82 0.80 0.19 0.18 0.24 0.18 0.18texas 0.62 0.32 0.77 ... 0.55 0.54 0.24 0.21 0.29 0.12 0.12. -
Consistent Kernel Mean Estimationfor Functions of Random Variables…
https://mlg.eng.cam.ac.uk/pub/pdf/SimSciToletal16.pdf13 Feb 2023: E[(. [µ̂hf(X) µhf(X)](z). )2]dz. Cn12c r=R‖f‖K. rs22e2re(s2d′)2/(4r) dr. Cn12ce(s2d. ′)24(R‖f‖K). r=R‖f‖K. ... Cn12ce(s2d. ′)24(R‖f‖K) e2(R‖f‖K)(R‖f‖K). s22. (for R ‖f‖K 1) (18) Cn12c(R‖f‖K). s22e2(R‖f‖K) (19). -
psb04.dvi
https://mlg.eng.cam.ac.uk/pub/pdf/DubHwaRan04a.pdf13 Feb 2023: HG beta. /?1& 5/, B1&0?; Q6=IRB.8+& 562+;,1&0K%/&0;-0&0,:U 8G&? <0O',.-<=5& ;(-<B.&,JB18G& ;L/ 24&?E6=IB. ... 6? <B16;)6;5 O? 18B%7 B.8B.8+&;-05/,.6; 6=I,.B?1/-CB./?<=54;GI 6? <B.6;VUL&0K%/&0;-0&5D<=24&05,6; -
gf2gp.dvi
https://mlg.eng.cam.ac.uk/zoubin/papers/gf2gp.pdf27 Jan 2023: 26. 24. 22. 20. 18. 16. α2α1. %(. 0.1. 0.08. 0.06. ... 150. 100. 50. 0 0.6. 0.8. 1. 28. 26. 24. 22. -
BIOINFORMATICS Vol. 20 no. 9 2004, pages 1361–1372DOI:…
https://mlg.eng.cam.ac.uk/zoubin/papers/Bioinformatics04rangel.pdf27 Jan 2023: Cells were collected in 300 µl ofRTL lysing solution (Qiagen) at the following times aftertreatment: 0, 2, 4, 6, 8, 18, 24, 32, 48, 72 h. ... Thecells used in this experiment were all expressing the T-cellreceptor (detected with anti CD3 antibodies) and -
Bayesian Gaussian Process Classificationwith the EM-EP…
https://mlg.eng.cam.ac.uk/zoubin/papers/KimGha06-PAMI.pdf27 Jan 2023: Itsgeneralized version which is convergent but slower hasbeen proposed [24]. 3.2 EP for Gaussian Process Classification. ... 00 0. CfJ. 24. 35; ð37Þ. where Cfj is a covariance matrix of latent values related to. -
Gaussian Process Regression Networks Andrew Gordon Wilson∗ David A.…
https://mlg.eng.cam.ac.uk/pub/pdf/WilKnoGha11.pdf13 Feb 2023: ȳ(x)i =k. E(Wik)E[fk] (24). cov(y(x))ij =k. [E(Wik)E(Wjk)var(fk) δij(var(Wik)E(f2k ))] δijσ2y (25). -
psb04.dvi
https://mlg.eng.cam.ac.uk/pub/pdf/DubHwaRanetal04.pdf13 Feb 2023: HG beta. /?1& 5/, B1&0?; Q6=IRB.8+& 562+;,1&0K%/&0;-0&0,:U 8G&? <0O',.-<=5& ;(-<B.&,JB18G& ;L/ 24&?E6=IB. ... 6? <B16;)6;5 O? 18B%7 B.8B.8+&;-05/,.6; 6=I,.B?1/-CB./?<=54;GI 6? <B.6;VUL&0K%/&0;-0&5D<=24&05,6; -
btc654.tex
https://mlg.eng.cam.ac.uk/zoubin/papers/Bioinformatics02raval.pdf27 Jan 2023: 0 226 0 98 310 0 14 322 0 214 24 0 1 23 0 2 4 0 21 16 0 9 24 0 115 205 51 20 265 0 11 ... 0 36 20 0 32 17 0 3518 28 23 0 28 0 23 23 0 28 27 0 24 28 0 2319 22 10 8 32 0 8 30 0 -
Curriculum Vitae Zoubin Ghahramani FRS CONTACT DETAILS Department:…
https://mlg.eng.cam.ac.uk/zoubin/fullcv.pdf27 Jan 2023: Curriculum Vitae. Zoubin Ghahramani FRS. CONTACT DETAILS. Department: Department of EngineeringUniversity of CambridgeTrumpington StreetCambridge CB2 1PZ, UK. Tel: 44 (0)1223 748 531Email: zoubin@eng.cam.ac.ukWWW: http://learning.eng.cam.ac.uk/zoubin -
~~~::'-"-~ ;~~ Cahiers de Psychologie Cognitive/~ Current…
https://mlg.eng.cam.ac.uk/zoubin/papers/KorGhaCPC.pdf27 Jan 2023: '-"-. ; Cahiers de Psychologie Cognitive/ Current Psychology of Cognition:;i; 2002, 21 (4-5), 537-564. ,'. A Bayesian view of motor adaptation. Alexander T. Korenberg and Zoubin Ghahramani. University College London, U.K. Abstract.The adaptability -
Bayesian nonparametrics andthe probabilistic approach to modelling…
https://mlg.eng.cam.ac.uk/pub/pdf/Gha12.pdf13 Feb 2023: Bayesian nonparametrics andthe probabilistic approach to modelling. Zoubin Ghahramani. Department of EngineeringUniversity of Cambridge, UK. zoubin@eng.cam.ac.ukhttp://mlg.eng.cam.ac.uk/zoubin. Modelling is fundamental to many fields of science and -
Bayesian Segmental Models withMultiple Sequence Alignment Profilesfor …
https://mlg.eng.cam.ac.uk/pub/pdf/ChuGhaPod06a.pdf13 Feb 2023: The existence of correlated side chain mutations in -heliceshas been well studied [20], [24]. -
� ��������� � ������ ��������� ������ ����! "����#��$&%…
https://mlg.eng.cam.ac.uk/pub/pdf/GoeRasToletal04.pdf13 Feb 2023: ÿ Y yz vOzNE EdpÒvkppwiA N yg|shdpYÓ+Xshi Ò ÔAy}yqtkwKqtshp. 8 16 24 32. -
paper.dvi
https://mlg.eng.cam.ac.uk/zoubin/papers/fhmmML.pdf27 Jan 2023: 0.67 9.81 2.55Exact 1.02 1.04 1.26 0.99Gibbs 2.21 0.91 2.50 0.87CFVA 1.24 1.50 1.50 1.53SVA -
PII: S0893-6080(02)00040-0
https://mlg.eng.cam.ac.uk/pub/pdf/UedGha02a.pdf13 Feb 2023: ðB23Þ. B.5. QðZlmÞ. Using Eqs. (6) and (24). QðZlmÞ / expfklog pðD; Zlw; m; S; W; b; mÞlQðm;S;W;blmÞg. -
ibp6.dvi
https://mlg.eng.cam.ac.uk/zoubin/papers/ibptr.pdf27 Jan 2023: B( αK , 1) =Γ( αK ). Γ(1 αK )= Kα , (24). -
http://ijr.sagepub.com/Robotics Research The International Journal of …
https://mlg.eng.cam.ac.uk/pub/pdf/BouAllKre04a.pdf13 Feb 2023: ξt (I At )ξt1 ̂t At µt1 Sx δt. (24)Hence, under this approximation the random variableξt isagain Gaussian distributed. ... Its mean is obtained by replacingξt andδt in eq. (24) by their respective means:. -
http://ijr.sagepub.com/Robotics Research The International Journal of …
https://mlg.eng.cam.ac.uk/pub/pdf/ThrLiuKol04a.pdf13 Feb 2023: ξt (I At )ξt1 ̂t At µt1 Sx δt. (24)Hence, under this approximation the random variableξt isagain Gaussian distributed. ... Its mean is obtained by replacingξt andδt in eq. (24) by their respective means:. -
paper.dvi
https://mlg.eng.cam.ac.uk/zoubin/papers/GhaGriSol06.pdf27 Jan 2023: E[A|X, Z] = (ZT Z σ2Xσ2A. I)1ZT X. (24). 1This simple toy example was inspired by the “shapes problem” in (Ghahramani, 1995);a larger scale example with real images -
Factorial Hidden Markov Models
https://mlg.eng.cam.ac.uk/pub/pdf/GhaJor97a.pdf13 Feb 2023: 2.50 0.87CFVA 1.24 1.50 1.50 1.53SVA 0.64 0.88 0.90 0.84. -
An Introduction to Variational Methods for Graphical Models
https://mlg.eng.cam.ac.uk/pub/pdf/JorGhaJaa99a.pdf13 Feb 2023: f (x) = maxλ{λT x f (λ)}, (23). where. f (λ) = maxx{λT x f (x)} (24). -
Bayesian Segmental Models withMultiple Sequence Alignment Profilesfor …
https://mlg.eng.cam.ac.uk/zoubin/papers/ieee-tcbb06.pdf27 Jan 2023: The existence of correlated side chain mutations in -heliceshas been well studied [20], [24]. -
Flexible Latent Variable Models for Multi-Task Learning Jian Zhang1,…
https://mlg.eng.cam.ac.uk/pub/pdf/ZhaGhaYan08.pdf13 Feb 2023: Flexible Latent Variable Models for Multi-Task. Learning. Jian Zhang1, Zoubin Ghahramani2,3, and Yiming Yang3. 1 Department of Statistics, Purdue University, West Lafayette, IN 47907jianzhan@stat.purdue.edu. 2 Department of Engineering, University -
Robust estimation of local genetic ancestry in admixed populations ...
https://mlg.eng.cam.ac.uk/pub/pdf/SohGhaXin12.pdf13 Feb 2023: Figure 10 about here.]. 24. Empirical analysis of HGDP data To illustrate our method on real data, we applied it. ... proportion of 0.24 in Native American ancestry. Although only one or two populations are. -
Variational Inference for the IndianBuffet Process Finale…
https://mlg.eng.cam.ac.uk/pub/pdf/DosMilVanTeh09b.pdf13 Feb 2023: 25 2973.7 -2.247. 5 163.24 -1.066. Finite Variational 10 767.1 -0.908. -
Unsupervised Learning∗ Zoubin Ghahramani† Gatsby Computational…
https://mlg.eng.cam.ac.uk/zoubin/papers/ul.pdf27 Jan 2023: The model can again be writtenas in (24):. P (s1:T , y1:T |θ) =T. ... HMMs can be extended by allowing a vector of discretestate variables, in an architecture known as a factorial HMM [24]. -
Metropolis Algorithms for Representative Subgraph Sampling Christian…
https://mlg.eng.cam.ac.uk/pub/pdf/HueBorKriGha08.pdf13 Feb 2023: 346 7 283 32HEP P H 34,546 420,899 24 846 61. -
Unsupervised Learning∗ Zoubin Ghahramani† Gatsby Computational…
https://mlg.eng.cam.ac.uk/zoubin/course05/ul.pdf27 Jan 2023: The model can again be writtenas in (24):. P (s1:T , y1:T |θ) =T. ... HMMs can be extended by allowing a vector of discretestate variables, in an architecture known as a factorial HMM [24]. -
Unsupervised Learning∗ Zoubin Ghahramani† Gatsby Computational…
https://mlg.eng.cam.ac.uk/zoubin/course04/ul.pdf27 Jan 2023: The model can again be writtenas in (24):. P (s1:T , y1:T |θ) =T. ... HMMs can be extended by allowing a vector of discretestate variables, in an architecture known as a factorial HMM [24]. -
LNAI 3176 - Unsupervised Learning
https://mlg.eng.cam.ac.uk/pub/pdf/Gha03a.pdf13 Feb 2023: p(x1:T , y1:T |θ) =T. t=1. p(xt|xt1, θ)p(yt|xt, θ) (24). In order words, the observations are assumed to have been generated fromthe hidden ... We assume that st can take discrete values in {1,. , K}.The model can again be written as in (24):. -
LETTER Communicated by Joris Mooij Model Reductions for Inference: ...
https://mlg.eng.cam.ac.uk/pub/pdf/EatGha13a.pdf13 Feb 2023: LETTER Communicated by Joris Mooij. Model Reductions for Inference: Generality of Pairwise,Binary, and Planar Factor Graphs. Frederik Eatonfrederik@ofb.netZoubin Ghahramanizoubin@eng.cam.ac.ukDepartment of Engineering, University of Cambridge, -
Proc. Valencia / ISBA 8th World Meeting on Bayesian ...
https://mlg.eng.cam.ac.uk/pub/pdf/GhaGriSol07.pdf13 Feb 2023: E[A|X, Z] = (ZT Z σ2Xσ2A. I)1ZT X. (24). 1This simple toy example was inspired by the “shapes problem” in (Ghahramani, 1995);a larger scale example with real images -
PII: S0893-6080(02)00040-0
https://mlg.eng.cam.ac.uk/zoubin/papers/UedaNeuralNetworks02.pdf27 Jan 2023: ðB23Þ. B.5. QðZlmÞ. Using Eqs. (6) and (24). QðZlmÞ / expfklog pðD; Zlw; m; S; W; b; mÞlQðm;S;W;blmÞg. -
LNCS 3355 - Analysis of Some Methods for Reduced Rank Gaussian…
https://mlg.eng.cam.ac.uk/pub/pdf/QuiRas05b.pdf13 Feb 2023: K1mm K. nm Z. Knmθi. ],. log |Q̃|σ2. =n m. σ2+ Tr [Zmm] ,. (24)where we have introduced Z Knm. [Knm Knm σ. 2 Kmm]1. -
doi:10.1016/j.neucom.2008.12.019
https://mlg.eng.cam.ac.uk/pub/pdf/DeiRasPet09.pdf13 Feb 2023: ARTICLE IN PRESS. Neurocomputing 72 (2009) 1508–1524. Contents lists available at ScienceDirect. Neurocomputing. 0925-23. doi:10.1. CorrTrumpi. E-m. journal homepage: www.elsevier.com/locate/neucom. Gaussian process dynamic programming. Marc Peter -
switch-ftp.dvi
https://mlg.eng.cam.ac.uk/zoubin/papers/switch-ftp.pdf27 Jan 2023: m) (23)q(m)t = exp 12T Yt C(m)X(m)t 0 R1 Yt C(m)X(m)t : (24)Here T is a temperature parameter, whi h is -
Max–Planck–Institut f ür biologische KybernetikMax Planck Institute…
https://mlg.eng.cam.ac.uk/pub/pdf/KusPfiCsaRas05.pdf13 Feb 2023: Max–Planck–Institut f ür biologische KybernetikMax Planck Institute for Biological Cybernetics. Technical Report No. 136. Approximate Inference forRobust Gaussian Process. Regression. Malte Kuss1, Tobias Pfingsten1,2, Lehel Csató1,Carl E. -
erice-top.dvi
https://mlg.eng.cam.ac.uk/zoubin/papers/varintro.pdf27 Jan 2023: we have:f(x) = max fTx f()g; (23)where f() = maxx fTx f(x)g (24)is the conjugate function.We have focused on linear bounds in this section, but ... 24 MICHAEL I. JORDAN ET AL.calculate the conjugate function. If the function is not convex or concave,then -
Bayesian Analysis (2006) 1, Number 4, pp. 793–832 Variational ...
https://mlg.eng.cam.ac.uk/zoubin/papers/BeaGha06.pdf27 Jan 2023: ln p(y |m)BIC = ln p(y | θ̂) d(m). 2ln n ln S (24). -
griffiths11a.dvi
https://mlg.eng.cam.ac.uk/pub/pdf/GriGha11.pdf13 Feb 2023: Journal of Machine Learning Research 12 (2011) 1185-1224 Submitted 3/10; Revised 3/11; Published 4/11. The Indian Buffet Process: An Introduction and Review. Thomas L. Griffiths TOM GRIFFITHS@BERKELEY.EDUDepartment of PsychologyUniversity of -
60 Denotational Validation of Higher-Order Bayesian Inference ADAM…
https://mlg.eng.cam.ac.uk/pub/pdf/SciKamVaketal18.pdf13 Feb 2023: 60. Denotational Validation of Higher-Order Bayesian Inference. ADAM ŚCIBIOR, University of Cambridge, England, UK and Max Planck Institute for Intelligent Systems,Germany. OHAD KAMMAR, University of Oxford, England, UKMATTHIJS VÁKÁR, University -
Journal of Machine Learning Research 10 (2009) 1187-1238 Submitted ...
https://mlg.eng.cam.ac.uk/pub/pdf/SilGha09.pdf13 Feb 2023: Journal of Machine Learning Research 10 (2009) 1187-1238 Submitted 4/08; Revised 11/08; Published 6/09. The Hidden Life of Latent Variables:Bayesian Learning with Mixed Graph Models. Ricardo Silva RICARDO@STATS.UCL.AC.UKDepartment of Statistical -
fit-epinions-svec.eps
https://mlg.eng.cam.ac.uk/pub/pdf/LesChaKleetal10.pdf13 Feb 2023: Journal of Machine Learning Research 11 (2010) 985-1042 Submitted 12/08; Revised 8/09; Published 2/10. Kronecker Graphs: An Approach to Modeling Networks. Jure Leskovec JURE@CS.STANFORD.EDUComputer Science DepartmentStanford UniversityStanford, CA -
Learning dynamic Bayesian networks
https://mlg.eng.cam.ac.uk/pub/pdf/Gha97a.pdf13 Feb 2023: e a d e r t o relevant t e x t s [41, 24, 19] for details. ... 21). (22) (23) (24) (25). where X and V are the prior m e a n and covariance o f t h e state, which are m o d e l -
Scalable Inference for StructuredGaussian Process Models Yunus…
https://mlg.eng.cam.ac.uk/pub/pdf/Saa11.pdf13 Feb 2023: 1.24). Note that the posterior over f is also an N-dimensional Gaussian as it is also con-. -
Generalised Bayesian Matrix Factorisation Models Shakir Mohamed St…
https://mlg.eng.cam.ac.uk/pub/pdf/Moh11.pdf13 Feb 2023: 24). The Hamiltonian can be seen as the log of the augmented distribution to be sampledfrom: p(θ, u|ψ) = p(θ|ψ)N(u|0, I). -
The Indian Buffet Process: Scalable Inference and Extensions Finale…
https://mlg.eng.cam.ac.uk/pub/pdf/Dos09.pdf13 Feb 2023: 24. 4 Accelerated Sampling in Conjugate Models 26. 4.1 Intuition. 27. ... poor modes early on. Chapter 3: Inference 24. Applied to the IBP, we can think of particle filtering as considering a num-.
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
Recent searches
Your search history is empty.