ICML Best Papers CollectingICML stands for International Conference on Machine Learning. ICML has now grown into an annual top-level international conference on machine learning hosted by the International Society for Machine Learning (IMLS).
International Conference on Machine Learning, (2019): 862-871
The most well known drawback of Gaussian processes regression is the computational cost of the exact calculation of these quantities, which scales as O N 3 in time and O N 2 in memory where N is the number of training examples
Cited by0BibtexViews256Links
0
0
ICML, (2018): 274-283
A phenomenon exhibited by certain defenses that makes standard gradient-based methods fail to generate adversarial examples
Cited by906BibtexViews184Links
0
0
international joint conference on artificial intelligence, (2018): 3150-3158
We argue that without a careful model of delayed outcomes, we cannot foresee the impact a fairness criterion would have if enforced as a constraint on a classification system
Cited by120BibtexViews197Links
0
0
arXiv: Learning, (2018)
As currently there does not seem to exist a reliable strategy to choose hyperparameters in the unsupervised learning of disentangled representations, we argue that future work should make the role of inductive biases and implicit and explicit supervision more explicit
Cited by72BibtexViews624Links
0
0
ICML, (2017)
We have discussed a variety of applications, from creating training-set attacks to debugging models and fixing datasets. Underlying each of these applications is a common tool, the influence function, which is based on a simple idea — we can better understand model behavior by lo...
Cited by646BibtexViews213Links
0
0
ICML, pp.190-198, (2014)
Arbitrary manipulations of the latent Dirichlet allocation model have to be introduced in order to adapt the topic model to a particular context; Zhao et al; Carman et al )
Cited by192BibtexViews159Links
0
0
ICML, (2013): 855-863
It may be viewed as a special case of a proximal minimization algorithm that uses Bregman divergences derived from submodular functions
Cited by92BibtexViews65Links
0
0
ICML, pp.597-605, (2013)
Algebraic geometry is a deep and fascinating field in mathematics, which deals with the structure of polynomial equations
Cited by28BibtexViews73Links
0
0
ICML, (2012)
We proposed and evaluated architectural improvements in these models resulting in PixelRNNs with up to 12 Long Short-Term Memory layers
Cited by853BibtexViews167Links
0
0
ICML, (2012)
In this paper we address the following question: “Can we approximately sample from a Bayesian posterior distribution if we are only allowed to touch a small mini-batch of data-items for every sample we generate?”
Cited by216BibtexViews84Links
0
0
ICML, pp.118-944, (2009)
Linear constraints on K preserve the global topology of the input graph
Cited by207BibtexViews41Links
0
0
ICML, pp.928-935, (2008)
We discuss how the runtime of SVM optimization should decrease as the size of the training data increases. We present theoretical and empirical results demonstrating how a simple subgradient descent approach indeed displays such behavior, at least for linear kernels.
Cited by256BibtexViews50Links
0
0
international conference on machine learning, pp.209-216, (2007)
In this paper, we present an information-theoretic approach to learning a Mahalanobis distance function. We formulate the problem as that of minimizing the differential relative entropy between two multivariate Gaussians under con- straints on the distance function. We express th...
Cited by2147BibtexViews76Links
0
0
ICML, pp.201-208, (2006)
We described two non-convex algorithms using ConCave Convex Procedure that bring marked scalability improvements over the corresponding convex approaches, namely for Support Vector Machines and TSVMs
Cited by334BibtexViews73Links
0
0
ICML, pp.377-384, (2005)
We present a Support Vector Method that can directly optimize a large class of performance measures like F1-score, Precision/Recall Breakeven Point, Precision at k, and ROCArea
Cited by901BibtexViews51Links
0
0
ICML, pp.49-56, (1999)
In order to demonstrate the computational complexity of incremental least-squares temporal difference learning algorithm, results were conducted with three different problem sizes: 14, 102 and 402 states
Cited by251BibtexViews45
0
0
Keywords
Classical GraphConcave-convex ProgrammingConnectivity AlgorithmConnectivity StructureContingency TableConventional ClassificationConvex ProgrammingDerive Regret BoundDifferential Relative EntropyDimensionality Reduction Algorithm
Authors
Shai Shalev-Shwartz
Paper 2
Percy Liang
Paper 1
Blake Shaw
Paper 1
Sajid M. Siddiqi
Paper 1
Nicholas Carlini
Paper 1
Olivier Teytaud
Paper 1
Amir Globerson
Paper 1
Pang Wei Koh
Paper 1
Aäron Van Den Oord
Paper 1