novikoff, kleinberg, strogatz: education of a model student
january 2012 by chl
"a dilemma faced by teachers, and increasingly by designers of educational software, is the trade-off between teaching new material and reviewing what has already been taught. complicating matters, review is useful only if it is neither too soon nor too late. moreover, different students need to review at different rates. we present a mathematical model that captures these issues in idealized form."
by:timothy-novikoff
by:steven-strogatz
by:jon-kleinberg
learning
paper
spaced-rep
from delicious
january 2012 by chl
a simpler approach to matrix completion
january 2012 by chl
"this paper provides the best bounds to date on the number of randomly sampled entries required to reconstruct an unknown low-rank matrix. [...] the reconstruction is accomplished by minimizing the nuclear norm, or sum of the singular values, of the hidden matrix subject to agreement with the provided entries. if the underlying matrix satisfies a certain incoherence condition, then the number of entries required is equal to a quadratic logarithmic factor times the number of parameters in the singular value decomposition. the proof of this assertion is short, self contained, and uses very elementary analysis. the novel techniques herein are based on recent work in quantum information theory."
quantum-info-theory
via:shivak
paper
matrix-completion
from delicious
january 2012 by chl
[1107.5728v2] the network of global corporate control
october 2011 by chl
'the structure of the control network of transnational corporations affects global market competition and financial stability. so far, only small national samples were studied and there was no appropriate methodology to assess control globally. we present the first investigation of the architecture of the international ownership network, along with the computation of the control held by each global player. we find that transnational corporations form a giant bow-tie structure and that a large portion of control flows to a small tightly-knit core of financial institutions. this core can be seen as an economic "super-entity" that raises new important issues both for researchers and policy makers.'
na
interlocking-directorates
via:cshalizi
paper
from delicious
october 2011 by chl
[1108.1791v1] why philosophers should care about computational complexity
august 2011 by chl
"one might think that, once we know something is computable, how efficiently it can be computed is a practical question with little further philosophical importance. in this essay, I offer a detailed case that one would be wrong."
complexity
paper
by:scott-aaronson
later
via:llimllib
computation
from delicious
august 2011 by chl
[1106.2429] efficient online learning via randomized rounding
june 2011 by chl
"most online algorithms used in machine learning today are based on variants of mirror descent or follow-the-leader. in this paper, we present an online algorithm based on a completely different approach, which combines "random playout" and randomized rounding of loss subgradients."
ml
online-learning
paper
later
via:shivak
from delicious
june 2011 by chl
scientific communication as sequential art
may 2011 by chl
watts/strogatz paper turned interactive by bret victor.
na
paper
sci-pub
dynamic-illustrations
interactive-documents
from delicious
may 2011 by chl
[1105.0902] modeling network evolution using graph motifs
may 2011 by chl
Modeling Network Evolution Using Graph Motifs - new paper, with Python code from @drewconway
na
paper
by:drew-conway
later
from delicious
may 2011 by chl
citeseerx — stop word location and identification for adaptive text recognition
january 2011 by chl
150 words cover 50% of typical english text. by scanning page images for likely occurrences of such words (using only word image width [!]) and then determining their identities through a word shape classifier, character prototypes can be extracted and fonts be learned.
font-learning
ocr
img-proc
by:tin-kam-ho
paper
from delicious
january 2011 by chl
[1011.3854] a probabilistic and ripless theory of compressed sensing
november 2010 by chl
"this paper introduces a simple and very general theory of compressive sensing. in this theory, the sensing mechanism simply selects sensing vectors independently at random from a probability distribution f [...]. we prove that if [f] obeys a simple incoherence property and an isotropy property, one can faithfully recover approximately sparse signals from a minimal number of noisy measurements. the novelty is that our recovery results do not require the restricted isometry property (rip) - they make use of a much weaker notion - or a random model for the signal."
compressed-sensing
via:ded_maxim
rip
paper
later
riplessness
from delicious
november 2010 by chl
[1008.4686] data analysis recipes: fitting a model to data
september 2010 by chl
"we go through the many considerations involved in fitting a model to data, using as an example the fit of a straight line to a set of points in a two-dimensional plane. standard weighted least-squares fitting is only appropriate when there is a dimension along which the data points have negligible uncertainties, and another along which all the uncertainties can be described by gaussians of known variance; these conditions are rarely met in practice. we consider cases of general, heterogeneous, and arbitrarily covariant two-dimensional uncertainties, and situations in which there are bad data (large outliers), unknown uncertainties, and unknown but expected intrinsic scatter in the linear relationship being fit. above all we emphasize the importance of having a "generative model" for the data [...]"
data-analysis
line-fitting
paper
by:david-hogg
by:dustin-lang
by:jo-bovy
from delicious
september 2010 by chl
[1006.3868] philosophy and the practice of bayesian statistics
june 2010 by chl
"a substantial school in the philosophy of science identifies bayesian inference with inductive inference and even rationality as such, and seems to be strengthened by the rise and practical success of bayesian statistics. we argue that the most successful forms of bayesian statistics do not actually support that particular philosophy but rather accord much better with sophisticated forms of hypothetico-deductivism."
hypothetico-deductivism
paper
by:andrew-gelman
by:cshalizi
stat
bayes
from delicious
june 2010 by chl
pregel
june 2010 by chl
The long-awaited Pregel paper from #SIGMOD2010 has been posted in ACM DL: #Hadoop #MapReduce
pregel
paper
google
na
#Hadoop
#MapReduce
#SIGMOD2010
Hadoop
MapReduce
SIGMOD2010
from delicious
june 2010 by chl
[0911.1824] community structure in time-dependent, multiscale, and multiplex networks
may 2010 by chl
"we developed a generalized framework of network quality functions that allowed us to study the community structure of arbitrary multislice networks, which are combinations of individual networks coupled through links that connect each node in one network slice to itself in other slices. this framework allows one to study community structure in a very general setting encompassing networks that evolve over time, have multiple types of links (multiplexity), and have multiple scales."
na
paper
later
from delicious
may 2010 by chl
[1003.5474] angle tree: nearest neighbor search in high dimensions with low intrinsic dimensionality
march 2010 by chl
'the key idea of our approach is to store the angle (the "dihedral angle") between the data region (which is a low dimensional manifold) and the random hyperplane that splits the region (the "splitter"). we show that the dihedral angle can be used to obtain a tight lower bound on the distance between the query point and any point on the opposite side of the splitter. this in turn can be used to efficiently prune the search space.' / '[...] the angle tree is the most efficient known indexing structure for nearest neighbor queries in terms of preprocessing and space usage while achieving high accuracy and fast search time.'
angle-tree
knn
via:shivak
paper
later
from delicious
march 2010 by chl
a comparison of user-generated and automatic graph layouts - microsoft research
november 2009 by chl
"our results demonstrate that the best of the user-generated layouts performed as well as or better than the physics-based layout. orthogonal and circular automatic layouts were found to be considerably less effective than either the physics-based layout or the best of the user-generated layouts." / cool idea.
graph-layout
graph-viz
paper
by:tim-dwyer
november 2009 by chl
dynamics of bayesian updating with dependent data and misspecified models
october 2009 by chl
for future reference.
paper
by:cshalizi
october 2009 by chl
[0908.2284] classification by set cover: the prototype vector machine
august 2009 by chl
"the method is compatible with any dissimilarity measure, making it amenable to situations in which the data are not embedded in an underlying feature space or in which using a non-euclidean metric is desirable. indeed, we demonstrate on the much studied zip code data how the pvm can reap the benefits of a problem-specific metric. in this example, the pvm outperforms the highly successful 1-nn with tangent distance, and does so retaining fewer than half of the data points."
ml
pvm
knn
paper
via:agray0
august 2009 by chl
learning compressed sensing
august 2009 by chl
"in this paper we ask: given a training set typical of the signals we wish to measure, what are the optimal set of linear projections for compressed sensing? we show that the optimal projections are in general not the principal components nor the independent components of the data, but rather a seemingly novel set of projections that capture what is still uncertain about the signal, given the training set. we also show that the projections onto the learned uncertain components may far outperform random projections. this is particularly true in the case of natural images,
where random projections have vanishingly small signal to noise
ratio as the number of pixels becomes large."
compressed-sensing
random-projections
uncertain-components
paper
by:yair-weiss
by:hyun-sung-chang
by:bill-freeman
filetype:pdf
media:document
where random projections have vanishingly small signal to noise
ratio as the number of pixels becomes large."
august 2009 by chl
related tags
#Hadoop ⊕ #MapReduce ⊕ #SIGMOD2010 ⊕ actorscript ⊕ ai ⊕ angle-tree ⊕ announcement ⊕ approx ⊕ approximate-closest-pairs ⊕ arx:q-bio.nc/0609008 ⊕ author:chl ⊕ author:earl ⊕ await ⊕ bagging ⊕ bayes ⊕ bcrypt ⊕ bing ⊕ bio ⊕ bio-inf ⊕ bio-net ⊕ bitmap-indices ⊕ blockbuster ⊕ bm25 ⊕ body-part-recognition ⊕ boltzmann-machines ⊕ boosting ⊕ bulk-reads ⊕ by:alexei-efros ⊕ by:alistair-moffat ⊕ by:andrew-gelman ⊕ by:andrew-moore ⊕ by:antonio-torralba ⊕ by:bill-freeman ⊕ by:carl-hewitt ⊕ by:cosma-shalizi ⊕ by:cshalizi ⊕ by:david-hogg ⊕ by:drew-conway ⊕ by:dustin-lang ⊕ by:edo-liberty ⊕ by:glinden ⊕ by:h-dieter-zeh ⊕ by:hyun-sung-chang ⊕ by:james-hays ⊕ by:jo-bovy ⊕ by:jon-kleinberg ⊕ by:justin-zobel ⊕ by:kristina-lisa-klinkner ⊕ by:léon-bottou ⊕ by:marcelo-camperi ⊕ by:mdreid ⊕ by:nir-ailon ⊕ by:peter-turney ⊕ by:purnamrita-sarkar ⊕ by:reginald-d.-smith ⊕ by:rob-fergus ⊕ by:rob-pike ⊕ by:robert-bell ⊕ by:scott-aaronson ⊕ by:steven-strogatz ⊕ by:ted-dunning ⊕ by:tim-dwyer ⊕ by:timothy-novikoff ⊕ by:tin-kam-ho ⊕ by:yair-weiss ⊕ by:yehuda-koren ⊕ cf ⊕ cga ⊕ click-chain-model ⊕ click-models ⊕ clique-matrices ⊕ clustering ⊕ code ⊕ coincidence ⊕ community-discovery ⊕ complexity ⊕ compressed-sensing ⊕ compression ⊕ computation ⊕ consciousness ⊕ contagion ⊕ crawling ⊕ crf ⊕ ctr-prediction ⊕ cv ⊕ cv? ⊕ cvpr-2008 ⊕ data ⊕ data-analysis ⊕ data-parallel ⊕ deep-learning ⊕ deep-nets ⊕ density-estimation ⊕ difference-map ⊕ dim-red ⊕ disk-geometry ⊕ dist ⊕ distributed-awareness? ⊕ distributional-semantics ⊕ divergences ⊕ diversity ⊕ dna ⊕ dnazip ⊕ drl ⊕ dunning-log-likelihood ⊕ dy-na ⊕ dy-net ⊕ dynamic-illustrations ⊕ ea ⊕ edge-bundling ⊕ em ⊕ email ⊕ ensemble-methods ⊕ epidemic-centrality ⊕ evolution ⊕ evolutionary-biology ⊕ extensions ⊕ facebook ⊕ filetype:pdf ⊕ fmm ⊕ fmri ⊕ fn-com ⊕ font-learning ⊕ fpga ⊕ ga ⊕ general-hilarity ⊕ geo ⊕ geometry ⊕ global-opt ⊕ google ⊕ google-chrome ⊕ gpgpu ⊕ gps ⊕ graph-layout ⊕ graph-viz ⊕ Hadoop ⊕ hamming-embedding ⊕ hashing ⊕ haskell ⊕ hdd ⊕ hmm ⊕ homophily ⊕ hypothetico-deductivism ⊕ im2gps ⊕ img-proc ⊕ img-search ⊕ indexing ⊕ info-geometry ⊕ info-sec ⊕ info-theory ⊕ information-forests ⊕ integrated-information-theory ⊕ interactive-documents ⊕ interlocking-directorates ⊕ internet ⊕ interpretability ⊕ io ⊕ ir ⊕ johnson-lindenstrauss ⊕ k) ⊕ k-means ⊕ k-tree ⊕ kd-trees ⊕ kdd-2009 ⊕ kernel-distance ⊕ kinect ⊕ knn ⊕ lambdarank ⊕ lang:de ⊕ latent-space-model ⊕ later ⊕ later? ⊕ lear ⊕ learning ⊕ likelihood-ratio-tests ⊕ lin-alg ⊕ line-fitting ⊕ ling ⊕ llr ⊕ lsa ⊕ lsh ⊕ MapReduce ⊕ matlab ⊕ matrix-approximation ⊕ matrix-completion ⊕ max-flow ⊕ media:document ⊕ media:pdf ⊕ message-passing ⊕ mf ⊕ microblogging ⊕ mine ⊕ ml ⊕ msft ⊕ msr ⊕ msr-accelerator ⊕ multi-relational-graphs ⊕ multi-scale ⊕ music ⊕ n-body ⊕ n-gram ⊕ na ⊕ neumaier's-clouds ⊕ neuro-sci ⊕ nflxprize ⊕ nips-2009 ⊕ nlp ⊕ nn ⊕ node-harvest ⊕ ocas ⊕ ocr ⊕ online-alg ⊕ online-learning ⊕ opt ⊕ p-boxes ⊕ p:d3c0v ⊕ p:m3p2r ⊕ pairwise-preferences ⊕ paper ⊖ paper-transistors ⊕ pathways ⊕ pca ⊕ physics ⊕ pkm ⊕ pmi-ir ⊕ poisson-intensity-reconstruction ⊕ possibility-distributions ⊕ preference-lists ⊕ pregel ⊕ prob ⊕ probability ⊕ probability-intervals ⊕ pvm ⊕ quantum-info-theory ⊕ query-classification ⊕ random-drift ⊕ random-forests ⊕ random-indexing ⊕ random-projections ⊕ random-sets ⊕ ranking ⊕ rec-sys ⊕ redundant-bit-vectors ⊕ ref+diff ⊕ refresh:1 ⊕ regression ⊕ reread:2007-01 ⊕ research-programme ⊕ review ⊕ rip ⊕ riplessness ⊕ scene-completion ⊕ scene-parsing ⊕ sci-pub ⊕ search ⊕ security ⊕ semantics ⊕ semi-supervised-learning ⊕ short-codes ⊕ SIGMOD2010 ⊕ sketching ⊕ sna ⊕ soc-net ⊕ sochastic-approximation ⊕ social-media ⊕ spaced-rep ⊕ sparse-coding ⊕ spectral-clustering ⊕ spectral-graph-theory ⊕ spiral-tap ⊕ sregexp ⊕ stat ⊕ storage ⊕ sum-products ⊕ summary ⊕ surf ⊕ surprise ⊕ svm ⊕ temporal-networks ⊕ texture-synthesis ⊕ tf-idf ⊕ thumbnailing ⊕ toefl ⊕ topic-models ⊕ tracking ⊕ traffic ⊕ transistors ⊕ trees ⊕ ts ⊕ twitter ⊕ uncertain-components ⊕ uncertainty ⊕ vanilla ⊕ vectors ⊕ via:adulau ⊕ via:agray0 ⊕ via:alstrup ⊕ via:arthegall ⊕ via:boomblitz? ⊕ via:csantos ⊕ via:cshalizi ⊕ via:ded_maxim ⊕ via:elzzup ⊕ via:gappy ⊕ via:glinden ⊕ via:herrmann ⊕ via:llimllib ⊕ via:nikete ⊕ via:ogrisel ⊕ via:shivak ⊕ via:simon.belak ⊕ via:wmf ⊕ via:wnpxrz ⊕ vocabulary ⊕ vsm ⊕ vxinsight ⊕ vxord ⊕ y! ⊕ year:2005 ⊕ yehuda-koren ⊕ ☆ ⊕Copy this bookmark: