In this paper, we propose a two-stage learning framework to perform self-supervised class-aware sounding object localization. First, we propose to learn robust object representations by aggregating the candidate sound localization results in the single source scenes. Then, class-aware object localization maps are generated in the cocktail-party scenarios by referring the pre-learned object knowledge, and the sounding objects are accordingly selected by matching audio and visual object category distributions, where the audiovisual consistency is viewed as the self-supervised signal. Experimental results in both realistic and synthesized cocktail-party videos demonstrate that our model is superior in filtering out silent objects and pointing out the location of sounding objects of different classes.

However, knowing which data to present during different stages of training is a challenging problem.

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. Identifying conditional independence of variables in graphical models is a key to finding tractable solutions, and faithfulness is a key condition of these relationships. The authors provide necessary and sufficient conditions for determining faithfulness in Gaussian graphical models (based on partitioning variables outside the conditioning set into two disjoint subsets), and show how this theoretical result can be translated into an algorithm for determining if a distribution is faithful. PROS: - Clear, well-written paper with illustrative examples - Addresses a relevant problem and provides a meaningful theoretical result - Provides a practical test for faithfulness in Gaussian graphical model CONS: - Theoretical result is restricted to Gaussian graphical models Quality: This paper provides addresses a theoretical problem (faithfulness in Gaussian graphical models). The claims are well-reasoned and the proofs support the claims. The resulting algorithm provides a useful test of faithfulness.

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

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

Learning with auxiliary tasks can improve the ability of a primary task to generalise. However, this comes at the cost of manually labelling auxiliary data.

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. The authors combine two recent advances in Gaussian processes, spectral mixture kernels [5], and scalable Gaussian processes for data in grids [14, 22 see below], in order to tackle applications with high amount of data points, like texture extrapolation, inpainting, and video extrapolation. The paper includes a thorough evaluation of the framework proposed, and comparisons against sparse GP methods, with general purpose covariance functions, and spectral mixture kernels. Quality The paper is technically sound. The framework proposed by the authors achieves outstanding results in the different applications studied in the paper.

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. Line 33: I don't think it is accurate to attribute the recent success of supervised neural nets on various applications to BP and dropout. Firstly, learning nets with gradient descent has been around a long time, and the key to its recent success has mostly been fast computers/GPUs, a wealth of labelled data, advances in understanding of how to make SGD work well (e.g. Techniques like dropout have also been useful in reducing overfitting, but are hardly the key missing ingredient to make these systems work well. Line 37: The claim that the lacklustre the results associated with unsupervised generative approaches is owed purely to their intractability issues is a strong and problematic one.