Colour vision has long fascinated scientists, who have sought to understand both the physiology of the mechanics of colour vision and the psychophysics of colour perception. We consider representations of colour in anatomically constrained convolutional deep neural networks. Following ideas from neuroscience, we classify cells in early layers into groups relating to their spectral and spatial functionality. We show the emergence of single and double opponent cells in our networks and characterise how the distribution of these cells changes under the constraint of a retinal bottleneck. Our experiments not only open up a new understanding of how deep networks process spatial and colour information, but also provide new tools to help understand the black box of deep learning. The code for all experiments is avaialable at \url{https://github.com/ecs-vlc/opponency}.

Classical supervised classification tasks search for a nonlinear mapping that maps each encoded feature directly to a probability mass over the labels. Such a learning framework typically lacks the intuition that encoded features from the same class tend to be similar and thus has little interpretability for the learned features. In this paper, we propose a novel supervised learning model named Supervised-Encoding Quantizer (SEQ). The SEQ applies a quantizer to cluster and classify the encoded features. We found that the quantizer provides an interpretable graph where each cluster in the graph represents a class of data samples that have a particular style. We also trained a decoder that can decode convex combinations of the encoded features from similar and different clusters and provide guidance on style transfer between sub-classes.

As one of the most ubiquitously applied unsupervised learning methods, clustering has also been known to have a few disadvantages. More specifically, parameters such as the number of clusters and neighborhood radius are what call the `unsupervised` nature of these algorithms into question. Moreover, the stochastic nature of a great number of these algorithms is also a considerable point of weakness. In order to address these issues, we propose DISCERN which can serve as an initialization algorithm for K-Means, finding suitable centroids that increase the performance of K-Means. Following that, the algorithm can estimate the number of clusters if need be. The algorithm does all of that, while maintaining complete robustness and returning the same results at each separate run. We ran experiments on the proposed method processing multiple datasets and the results show its undeniable superiority in terms of results, computational time and robustness when compared to the randomized K-Means and K-Means++ initialization. In addition, the superiority in estimating the number of clusters is also discussed and we prove the lower complexity when compared to methods such as the elbow and silhouette methods in estimating the number of clusters.

A variety of papers have been recently produced on "robustifying " Deep Neural Networks (DNNs), particularly to adversarial Test-Time Evasion (TTE) attacks [14, 15, 13]. We discuss some of this work in Sections III.A and IV.A of [9 ] and argue for the need for TTE-attack detection [8] for robustness . In this note, we derive a local class purity result under the assumption of Lipschitz continuity, discuss Lipschitz margin training, and define an associated p-value. Estimation of the Lipschitz parameter for a given DNN is disc ussed in, e.g., [12, 14, 16, 4].

Time Series Classification (TSC) has been an important and challenging task in data mining, especially on multivariate time series and multi-view time series data sets. Meanwhile, transfer learning has been widely applied in computer vision and natural language processing applications to improve deep neural network's generalization capabilities. However, very few previous works applied transfer learning framework to time series mining problems. Particularly, the technique of measuring similarities between source domain and target domain based on dynamic representation such as density estimation with importance sampling has never been combined with transfer learning framework. In this paper, we first proposed a general adaptive transfer learning framework for multi-view time series data, which shows strong ability in storing inter-view importance value in the process of knowledge transfer. Next, we represented inter-view importance through some time series similarity measurements and approximated the posterior distribution in latent space for the importance sampling via density estimation techniques. We then computed the matrix norm of sampled importance value, which controls the degree of knowledge transfer in pre-training process. We further evaluated our work, applied it to many other time series classification tasks, and observed that our architecture maintained desirable generalization ability. Finally, we concluded that our framework could be adapted with deep learning techniques to receive significant model performance improvements.

This paper considers a semi-supervised learning framework for weakly labeled polyphonic sound event detection problems for the DCASE 2019 challenge's task4 by combining both the tri-training and adversarial learning. The goal of the task4 is to detect onsets and offsets of multiple sound events in a single audio clip. The entire dataset consists of the synthetic data with a strong label (sound event labels with boundaries) and real data with weakly labeled (sound event labels) and unlabeled dataset. Given this dataset, we apply the tri-training where two different classifiers are used to obtain pseudo labels on the weakly labeled and unlabeled dataset, and the final classifier is trained using the strongly labeled dataset and weakly/unlabeled dataset with pseudo labels. Also, we apply the adversarial learning to reduce the domain gap between the real and synthetic dataset. We evaluated our learning framework using the validation set of the task4 dataset, and in the experiments, our learning framework shows a considerable performance improvement over the baseline model.

Reanalysis datasets combining numerical physics models and limited observations to generate a synthesised estimate of variables in an Earth system, are prone to biases against ground truth. Biases identified with the NASA Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2) aerosol optical depth (AOD) dataset, against the Aerosol Robotic Network (AERONET) ground measurements in previous studies, motivated the development of a deep learning based AOD prediction model globally. This study combines a convolutional neural network (CNN) with MERRA-2, tested against all AERONET sites. The new hybrid CNN-based model provides better estimates validated versus AERONET ground truth, than only using MERRA-2 reanalysis.

In this paper, we present an acoustic scene classification framework based on a large-margin factorized convolutional neural network (CNN). We adopt the factorized CNN to learn the patterns in the time-frequency domain by factorizing the 2D kernel into two separate 1D kernels. The factorized kernel leads to learn the main component of two patterns: the long-term ambient and short-term event sounds which are the key patterns of the audio scene classification. In training our model, we consider the loss function based on the triplet sampling such that the same audio scene samples from different environments are minimized, and simultaneously the different audio scene samples are maximized. With this loss function, the samples from the same audio scene are clustered independently of the environment, and thus we can get the classifier with better generalization ability in an unseen environment. We evaluated our audio scene classification framework using the dataset of the DCASE challenge 2019 task1A. Experimental results show that the proposed algorithm improves the performance of the baseline network and reduces the number of parameters to one third. Furthermore, the performance gain is higher on unseen data, and it shows that the proposed algorithm has better generalization ability.

Data samples collected for training machine learning models are typically assumed to be independent and identically distributed (iid). Recent research has demonstrated that this assumption can be problematic as it simplifies the manifold of structured data. This has motivated different research areas such as data poisoning, model improvement, and explanation of machine learning models. In this work, we study the influence of a sample on determining the intrinsic topological features of its underlying manifold. We propose the Shapley Homology framework, which provides a quantitative metric for the influence of a sample of the homology of a simplicial complex. By interpreting the influence as a probability measure, we further define an entropy which reflects the complexity of the data manifold. Our empirical studies show that when using the 0-dimensional homology, on neighboring graphs, samples with higher influence scores have more impact on the accuracy of neural networks for determining the graph connectivity and on several regular grammars whose higher entropy values imply more difficulty in being learned.

Due in part to the growing sources of data about past sequences of decisions and their outcomes - from marketing to energy management to healthcare - there is increasing interest in developing accurate and efficient algorithms for off-policy policy evaluation. For Markov Decision Processes, this problem was addressed (Precup et al., 2000) early on by importance sampling (IS)(Rubinstein, 1981), a method prone to large variance due to rare events (Glynn, 1994; L'Ecuyer et al., 2009). The per-decision importance sampling estimator of Precup et al. (2000) tries to mitigate this problem by leveraging the temporal structure - earlier rewards cannot depend on later decisions - of the domain. While neither importance sampling (IS) nor per-decision IS (PDIS) assumes the underlying domain is Markov, more recently, a new class of estimators (Hallak and Mannor, 2017; Liu et al., 2018; Gelada and Bellemare, 2019) has been proposed that leverages the Markovian structure. In particular, these approaches propose performing importance sampling over the stationary state-action distributions induced by the corresponding Markov chain for a particular policy. By avoiding the explicit accumulation of likelihood ratios along the trajectories, it is hypothesized that such ratios of stationary distributions could substantially reduce the variance of the resulting estimator, thereby overcoming the "curse of horizon" (Liu et al., 2018) plaguing off-policy evaluation. The recent flurry of empirical results shows significant performance improvements over the alternative methods on a variety of simulation domains. Yet so far there has not been a formal analysis of the accuracy of IS, PDIS, and stationary state-action IS which will strengthen our understanding of their properties, benefits and limitations.