First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. Summary: This paper studies the Principal Component Analysis (PCA) for large tensors of arbitrary order k under a single-spike model. Solving tensor PCA exactly is in general NP hard. Given a completely observed rank-one symmetric tensor, this paper provides conditions under which one can reliably estimate the unknown unit vector. Specifically, the paper gives conditions on signal-to-noise ratio under several scenarios which allow one to estimate the solution reliably. For the maximum-likelihood estimator (MLE), the authors show that in an ideal case with unbounded computational resources, the MLE is successful with high probability if the signal-to-noise ration is above sqrt(k.log(k))(1+o(1))

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. This paper studies a planted partition model for random m-uniform hypergraphs, and proves the consistency of a natural generalization of spectral clustering. The hypergraph adjacency tensor is (mode-1) flattened to a matrix, from which a normalized Laplacian matrix is formed and the standard spectral partitioning is then applied. The striking feature of the analysis is that the rate of convergence improves as m increases, provided that the number of partitions is small. Some experiments on both synthetic and application derived data are reported, and the proposed method is shown to be relatively effective, especially given its simplicity. The model is well-motivated by applications in computer vision and likely elsewhere.

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. This paper defines a joint generative model of an image and its annotated text which is used to learn a bit vector representation for large scale image retrieval. An Indian Buffet Process is used to learn the length of the bit vector. The method is compared favourably to several widely used techniques. Quality ======= It is good to see a retrieval paper constructed around a well-defined probabilistic model.

Neural Information Processing Systems http://nips.cc/

Although optimization is the longstanding algorithmic backbone of machine learning, new models still require the time-consuming implementation of new solvers. As a result, there are thousands of implementations of optimization algorithms for machine learning problems.

Q2: Please summarize your review in 1-2 sentences very nice method, important application, could benefit from more explicit comparisons and demonstration of scalability.

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. The paper is concerned with Monte Carlo sampling based on the discretisation of SDEs. This is a particularly topical subject since there has been some interest lately in such techniques due to the fact that they allow for the use of stochastic gradients which are particularly appealing in some big data settings since they allow one to run algorithms with only partial evaluation of the likelihood/energy function. The paper is particularly well written and pedagogical. In additional it clarifies earlier contributions and provides a rigorous overview of the main results useful in this emerging area.

"NIPS Neural Information Processing Systems 8-11th December 2014, Montreal, Canada",,, "Paper ID:","406" "Title:","Learning Mixed Multinomial Logit Model from Ordinal Data" Current Reviews First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. Summary: This paper extends the classic MultiNomial Logit (MNL) choice model to a general family of choice models named Mixed MNL, which can be seen as a parametric class of distributions over permutations (e.g., permutations of items according to user preference). The main contributions of the paper are (1) to identify sufficient conditions under which a mixed MNL can be learnt, and (2) to propose a two-phase algorithm to learn the proposed mixed MNL models in an efficient manner. Part of the interesting theoretical results shows that the model with r components can be learnt with sample size being polynomially in n (number of items of interest) and r (number of components). Quality: The problem choice modeling studied in this paper is a fundamental and critical problem to the social choice community, and the proposed model and algorithm for this problem are certainly of interest to the machine learning community.

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. This paper provides two algorithms based on the soft-thresholding method for estimating a penalized pseudo-likelihood graphical model. The coordinate-wise method seems a practical improvement of the current CONCORD method. Overall, this is a worthwhile addition to a booming literature on this issue. The method has computational complexity O(sp2), but clearly also depends on the starting value.

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. This paper proposes a probabilistic approach for learning the assignment of exercises to skills from student data, where student knowledge changes while exercises are being solved; the model also estimates the student knowledge while estimating the skill assignments. The paper uses a weighted CRP to model the assignment, incorporating expert labelings through the weighting. In simulation, the method recovers skill labelings with high accuracy, with little dependence on the expert labels, and across several datasets, the paper finds that skill labelings from this method result in higher prediction accuracy than other approaches. Overall, I found the paper to be clear and the proposed model is a relatively novel extension of existing methods.