Daniel Povey, the main developer of the widely used open-source speech recognition toolkit Kaldi, tweeted today that he is likely joining Chinese smartphone giant Xiaomi at its Beijing headquarters to work on a next generation "PyTorch-y Kaldi." I am very close to signing an agreement to work for Xiaomi in Beijing. Would leave before end of 2019, and would hire a small team there to work on next-gen PyTorch-y' Kaldi. Povey is a leader in voice recognition research, known for his contributions to speech recognition and language processing technologies. He and other researchers first created Kaldi as part of a Johns Hopkins University workshop in 2009.

As the world enters the age of the fourth industrial revolution, marked by accelerating innovation and the adoption of automation, the future of work is a fundamental question for the Middle East. While some jobs will be lost and others will be created, nearly all jobs will be transformed. The new reality is one in which 45 per cent of jobs will be automatable by 2030. The automation potential will vary across sectors: jobs requiring repetitive routine work such as manufacturing, warehousing and transportation will see more than 50 per cent of its work done by smart devices. Jobs that require emotional intelligence and creativity such as the arts, health care and entertainment will only see a 29 to 37 per cent automation rate.

Sebastian Castro demonstrates an example of controlling humanoid robot locomotion using deep reinforcement learning, specifically the Deep Deterministic Policy Gradient (DDPG) algorithm. The robot is simulated using Simscape Multibody, while training the control policy is done using Reinforcement Learning Toolbox . In this video, Sebastian outlines the setup, training, and evaluation of reinforcement learning with Simulink models. First, he introduces how to choose states, actions, and a reward function for the reinforcement learning problem. Then he describes the neural network structure and training algorithm parameters.

Ernesto Estrada is ARAID researcher at the Institute of Mathematics and Applications (IUMA) at the University of Zaragoza since January 2019. Before he was the Chair of Complexity Science at the University of Strathclyde in Glasgow. He works on the mathematics of networks where he has published more than 200 papers which have received more than 12,500 citations, and his h-index is 59. He is SIAM Fellow, Member of the Academy of Sciences of Latin America, and was a recipient of the Wolfson Research Merit Award of the Royal Society of London among other distinctions. He is the Editor in Chief of the Journal of Complex Networks (Oxford University Press), and Associate Editor of SIAM Journal of Applied Mathematics and of Proceedings of the Royal Society A. He has given plenary talks at many international conferences in applied mathematics and on network sciences, and he is frequently a lecturer at major international schools on these topics.

Recent work by Brock et al. (2018) suggests that Generative Adversarial Networks (GANs) benefit disproportionately from large mini-batch sizes. Unfortunately, using large batches is slow and expensive on conventional hardware. Thus, it would be nice if we could generate batches that were effectively large though actually small. In this work, we propose a method to do this, inspired by the use of Coreset-selection in active learning. When training a GAN, we draw a large batch of samples from the prior and then compress that batch using Coreset-selection. To create effectively large batches of 'real' images, we create a cached dataset of Inception activations of each training image, randomly project them down to a smaller dimension, and then use Coreset-selection on those projected activations at training time. We conduct experiments showing that this technique substantially reduces training time and memory usage for modern GAN variants, that it reduces the fraction of dropped modes in a synthetic dataset, and that it allows GANs to reach a new state of the art in anomaly detection.

Our research aims at developing intelligent systems to reduce the transportation-related energy expenditure of a large city by influencing individual behavior. We introduce Copter - an intelligent travel assistant that evaluates multi-modal travel alternatives to find a plan that is acceptable to a person given their context and preferences. We propose a formulation for acceptable planning that brings together ideas from AI, machine learning, and economics. This formulation has been incorporated in Copter that produces acceptable plans in real-time. We adopt a novel empirical evaluation framework that combines human decision data with a high fidelity multi-modal transportation simulation to demonstrate a 4% energy reduction and 20% delay reduction in a realistic deployment scenario in Los Angeles, California, USA. This article is part of the special track on AI and Society.

Ensembles, as a widely used and effective technique in the machine learning community, succeed within a key element--"diversity." The relationship between diversity and generalization, unfortunately, is not entirely understood and remains an open research issue. To reveal the effect of diversity on the generalization of classification ensembles, we investigate three issues on diversity, i.e., the measurement of diversity, the relationship between the proposed diversity and generalization error, and the utilization of this relationship for ensemble pruning. In the diversity measurement, we measure diversity by error decomposition inspired by regression ensembles, which decomposes the error of classification ensembles into accuracy and diversity. Then we formulate the relationship between the measured diversity and ensemble performance through the theorem of margin and generalization, and observe that the generalization error is reduced effectively only when the measured diversity is increased in a few specific ranges, while in other ranges larger diversity is less beneficial to increase generalization of an ensemble. Besides, we propose a pruning method based on diversity management to utilize this relationship, which could increase diversity appropriately and shrink the size of the ensemble with non-decreasing performance. The experiments validate the effectiveness of this proposed relationship between the proposed diversity and the ensemble generalization error.

We propose a heuristically modified FP-Tree for ontology learning from text. Unlike previous research, for concept extraction, we use a regular expression parser approach widely adopted in compiler construction, i.e., deterministic finite automata (DFA). Thus, the concepts are extracted from unstructured documents. For ontology learning, we use a frequent pattern mining approach and employ a rule mining heuristic function to enhance its quality. This process does not rely on predefined lexico-syntactic patterns, thus, it is applicable for different subjects. We employ the ontology in a question-answering system for students' content-related questions. For validation, we used textbook questions/answers and questions from online course forums. Subject experts rated the quality of the system's answers on a subset of questions and their ratings were used to identify the most appropriate automatic semantic text similarity metric to use as a validation metric for all answers. The Latent Semantic Analysis was identified as the closest to the experts' ratings. We compared the use of our ontology with the use of Text2Onto for the question-answering system and found that with our ontology 80% of the questions were answered, while with Text2Onto only 28.4% were answered, thanks to the finer grained hierarchy our approach is able to produce.

We study a variant of decision-theoretic online learning in which the set of experts that are available to Learner can shrink over time. This is a restricted version of the well-studied sleeping experts problem, itself a generalization of the fundamental game of prediction with expert advice. Similar to many works in this direction, our benchmark is the ranking regret. Various results suggest that achieving optimal regret in the fully adversarial sleeping experts problem is computationally hard. This motivates our relaxation where any expert that goes to sleep will never again wake up. We call this setting "dying experts" and study it in two different cases: the case where the learner knows the order in which the experts will die and the case where the learner does not. In both cases, we provide matching upper and lower bounds on the ranking regret in the fully adversarial setting. Furthermore, we present new, computationally efficient algorithms that obtain our optimal upper bounds.

Model-agnostic meta-learners aim to acquire meta-learned parameters from similar tasks to adapt to novel tasks from the same distribution with few gradient updates. With the flexibility in the choice of models, those frameworks demonstrate appealing performance on a variety of domains such as few-shot image classification and reinforcement learning. However, one important limitation of such frameworks is that they seek a common initialization shared across the entire task distribution, substantially limiting the diversity of the task distributions that they are able to learn from. In this paper, we augment MAML with the capability to identify the mode of tasks sampled from a multimodal task distribution and adapt quickly through gradient updates. Specifically, we propose a multimodal MAML (MMAML) framework, which is able to modulate its meta-learned prior parameters according to the identified mode, allowing more efficient fast adaptation. We evaluate the proposed model on a diverse set of few-shot learning tasks, including regression, image classification, and reinforcement learning. The results not only demonstrate the effectiveness of our model in modulating the meta-learned prior in response to the characteristics of tasks but also show that training on a multimodal distribution can produce an improvement over unimodal training.