By and large the process of learning concepts that are embedded in time is regarded as quite a mature research topic. Hidden Markov models, recurrent neural networks are, amongst others, successful approaches to learning from temporal data. In this paper, we claim that the dominant approach minimizing appropriate risk functions defined over time by classic stochastic gradient might miss the deep interpretation of time given in other fields like physics. We show that a recent reformulation of learning according to the principle of Least Cognitive Action is better suited whenever time is involved in learning. The principle gives rise to a learning process that is driven by differential equations, that can somehow descrive the process within the same framework as other laws of nature.

We propose a new graph kernel for graph classification and comparison using Ollivier Ricci curvature. The Ricci curvature of an edge in a graph describes the connectivity in the local neighborhood. An edge in a densely connected neighborhood has positive curvature and an edge serving as a local bridge has negative curvature. We use the edge curvature distribution to form a graph kernel which is then used to compare and cluster graphs. The curvature kernel uses purely the graph topology and thereby works for settings when node attributes are not available.

To make music composition more approachable, we designed the first AI-powered Google Doodle, the Bach Doodle, where users can create their own melody and have it harmonized by a machine learning model Coconet (Huang et al., 2017) in the style of Bach. For users to input melodies, we designed a simplified sheet-music based interface. To support an interactive experience at scale, we re-implemented Coconet in TensorFlow.js (Smilkov et al., 2019) to run in the browser and reduced its runtime from 40s to 2s by adopting dilated depth-wise separable convolutions and fusing operations. We also reduced the model download size to approximately 400KB through post-training weight quantization. We calibrated a speed test based on partial model evaluation time to determine if the harmonization request should be performed locally or sent to remote TPU servers. In three days, people spent 350 years worth of time playing with the Bach Doodle, and Coconet received more than 55 million queries. Users could choose to rate their compositions and contribute them to a public dataset, which we are releasing with this paper. We hope that the community finds this dataset useful for applications ranging from ethnomusicological studies, to music education, to improving machine learning models.

In this paper, we develop and explore deep anomaly detection techniques based on the capsule network (CapsNet) for image data. Being able to encoding intrinsic spatial relationship between parts and a whole, CapsNet has been applied as both a classifier and deep autoencoder. This inspires us to design a prediction-probability-based and a reconstruction-error-based normality score functions for evaluating the "outlierness" of unseen images. Our results on three datasets demonstrate that the prediction-probability-based method performs consistently well, while the reconstruction-error-based approach is relatively sensitive to the similarity between labeled and unlabeled images. Furthermore, both of the CapsNet-based methods outperform the principled benchmark methods in many cases.

Although deep neural networks have been widely applied in various domains [19, 25, 27], usually its parameters are learned via the principle of maximum likelihood, hence its success crucially hinges on the availability of large scale datasets. When training rich models on small datasets, explicit regularization techniques are crucial to alleviate overfitting. Previous works have explored various regularization [39] and data augmentation [19, 38] techniques to learn diversified representations. In this paper, we look into an alternative direction by proposing an adaptive and data-dependent regularization method to encourage neurons of the same layer to share statistical strength. The goal of our method is to prevent overfitting when training (large) networks on small dataset. Our key insight stems from the famous argument by Efron [8] in the literature of the empirical Bayes method: It is beneficial to learn from the experience of others. From an algorithmic perspective, we argue that the connection weights of neurons in the same layer (row/column vectors of the weight matrix) will be correlated with each other through the backpropagation learning. Hence, by learning the correlations of the weight matrix, a neuron can "borrow statistical strength" from other neurons in the same layer.

We consider the weakly supervised binary classification problem where the labels are randomly flipped with probability $1- {\alpha}$. Although there exist numerous algorithms for this problem, it remains theoretically unexplored how the statistical accuracies and computational efficiency of these algorithms depend on the degree of supervision, which is quantified by ${\alpha}$. In this paper, we characterize the effect of ${\alpha}$ by establishing the information-theoretic and computational boundaries, namely, the minimax-optimal statistical accuracy that can be achieved by all algorithms, and polynomial-time algorithms under an oracle computational model. For small ${\alpha}$, our result shows a gap between these two boundaries, which represents the computational price of achieving the information-theoretic boundary due to the lack of supervision. Interestingly, we also show that this gap narrows as ${\alpha}$ increases. In other words, having more supervision, i.e., more correct labels, not only improves the optimal statistical accuracy as expected, but also enhances the computational efficiency for achieving such accuracy.

In this paper we describe a new algorithm called Fast Adaptive Sequencing Technique (FAST) for maximizing a monotone submodular function under a cardinality constraint $k$ whose approximation ratio is arbitrarily close to $1-1/e$, is $O(\log(n) \log^2(\log k))$ adaptive, and uses a total of $O(n \log\log(k))$ queries. Recent algorithms have comparable guarantees in terms of asymptotic worst case analysis, but their actual number of rounds and query complexity depend on very large constants and polynomials in terms of precision and confidence, making them impractical for large data sets. Our main contribution is a design that is extremely efficient both in terms of its non-asymptotic worst case query complexity and number of rounds, and in terms of its practical runtime. We show that this algorithm outperforms any algorithm for submodular maximization we are aware of, including hyper-optimized parallel versions of state-of-the-art serial algorithms, by running experiments on large data sets. These experiments show FAST is orders of magnitude faster than the state-of-the-art.

A standard way to tackle the challenging task of learning from high-dimensional data is to exploit its underlying low-dimensional structure. Union of Subspaces (UoS) is a popular and powerful model to describe such structure which assumes that the data lies in the union of a collection of low-dimensional subspaces. Extracting useful information from UoS structure of data has become the task of the newly-emerged field of subspace learning. In this paper, we investigate how random projection, an efficient and commonly-used method for dimensionality reduction, distorts the UoS structure of data. Here the fine details of UoS structure are described in terms of canonical angles (also known as principal angles) between subspaces, which is a well-known characterization for relative subspace positions by a sequence of angles. It is proved that random projection with the so-called Johnson-Lindenstrauss (JL) property approximately preserves canonical angles between subspaces. As canonical angles completely determine the relative position of subspaces, our result indicates that random projection approximately preserves structure of a union of subspaces. Inspired by this result, we propose in this paper the framework of Compressed Subspace Learning (CSL), which enables to extract useful information from the UoS structure of data in a greatly reduced dimension and has the advantage of lower computational cost and memory requirements. We demonstrate the effectiveness of CSL in various subspace-related tasks such as subspace visualization, active subspace detection, and subspace clustering.

We study two time-scale linear stochastic approximation algorithms, which can be used to model well-known reinforcement learning algorithms such as GTD, GTD2, and TDC. We present finite-time performance bounds for the case where the learning rate is fixed. The key idea in obtaining these bounds is to use a Lyapunov function motivated by singular perturbation theory for linear differential equations. We use the bound to design an adaptive learning rate scheme which significantly improves the convergence rate over the known optimal polynomial decay rule in our experiments, and can be used to potentially improve the performance of any other schedule where the learning rate is changed at pre-determined time instants.

We present some results concerning the dialogue behavior and inferred sentiment of a group of older adults interacting with a computer-based avatar. Our avatar is unique in its ability to hold natural dialogues on a wide range of everyday topics---27 topics in three groups, developed with the help of gerontologists. The three groups vary in ``degrees of intimacy", and as such in degrees of difficulty for the user. Each participant interacted with the avatar for 7-9 sessions over a period of 3-4 weeks; analysis of the dialogues reveals correlations such as greater verbosity for more difficult topics, increasing verbosity with successive sessions, especially for more difficult topics, stronger sentiment on topics concerned with life goals rather than routine activities, and stronger self-disclosure for more intimate topics. In addition to their intrinsic interest, these results also reflect positively on the sophistication of our dialogue system.