During sleep and awake rest, the hippocampus replays sequences of place cells that have been activated during prior experiences. These have been interpreted as a memory consolidation process, but recent results suggest a possible interpretation in terms of reinforcement learning. The Dyna reinforcement learning algorithms use off-line replays to improve learning. Under limited replay budget, a prioritized sweeping approach, which requires a model of the transitions to the predecessors, can be used to improve performance. We investigate whether such algorithms can explain the experimentally observed replays. We propose a neural network version of prioritized sweeping Q-learning, for which we developed a growing multiple expert algorithm, able to cope with multiple predecessors. The resulting architecture is able to improve the learning of simulated agents confronted to a navigation task. We predict that, in animals, learning the world model should occur during rest periods, and that the corresponding replays should be shuffled.

The discovery of time series motifs has emerged as one of the most useful primitives in time series data mining. Researchers have shown its utility for exploratory data mining, summarization, visualization, segmentation, classification, clustering, and rule discovery. Although there has been more than a decade of extensive research, there is still no technique to allow the discovery of time series motifs in the presence of missing data, despite the welldocumented ubiquity of missing data in scientific, industrial, and medical datasets. In this work, we introduce a technique for motif discovery in the presence of missing data. We formally prove that our method is admissible, producing no false negatives. We also show that our method can "piggyback" off the fastest known motif discovery method with a small constant factor time/space overhead. We will demonstrate our approach on diverse datasets with varying amounts of missing data.

Existing multi-agent reinforcement learning methods are limited typically to a small number of agents. When the agent number increases largely, the learning becomes intractable due to the curse of the dimensionality and the exponential growth of user interactions. In this paper, we present Mean Field Reinforcement Learning where the interactions within the population of agents are approximated by those between a single agent and the average effect from the overall population or neighboring agents; the interplay between the two entities is mutually reinforced: the learning of the individual agent's optimal policy depends on the dynamics of the population, while the dynamics of the population change according to the collective patterns of the individual policies. We develop practical mean field Q-learning and mean field Actor-Critic algorithms and analyze the convergence of the solution. Experiments on resource allocation, Ising model estimation, and battle game tasks verify the learning effectiveness of our mean field approaches in handling many-agent interactions in population.

Noisy observations coupled with nonlinear dynamics pose one of the biggest challenges in robot motion planning. By decomposing the nonlinear dynamics into a discrete set of local dynamics models, hybrid dynamics provide a natural way to model nonlinear dynamics, especially in systems with sudden "jumps" in the dynamics, due to factors such as contacts. We propose a hierarchical POMDP planner that develops locally optimal motion plans for hybrid dynamics models. The hierarchical planner first develops a high-level motion plan to sequence the local dynamics models to be visited. The high-level plan is then converted into a detailed cost-optimized continuous state plan. This hierarchical planning approach results in a decomposition of the POMDP planning problem into smaller sub-parts that can be solved with significantly lower computational costs. The ability to sequence the visitation of local dynamics models also provides a powerful way to leverage the hybrid dynamics to reduce state uncertainty. We evaluate the proposed planner for two navigation and localization tasks in simulated domains, as well as an assembly task with a real robotic manipulator.

We identify obfuscated gradients, a kind of gradient masking, as a phenomenon that leads to a false sense of security in defenses against adversarial examples. While defenses that cause obfuscated gradients appear to defeat iterative optimization-based attacks, we find defenses relying on this effect can be circumvented. For each of the three types of obfuscated gradients we discover, we describe characteristic behaviors of defenses exhibiting this effect and develop attack techniques to overcome it. In a case study, examining non-certified white-box-secure defenses at ICLR 2018, we find obfuscated gradients are a common occurrence, with 7 of 8 defenses relying on obfuscated gradients. Our new attacks successfully circumvent 6 completely and 1 partially.

ABSTRACT Training a conventional automatic speech recognition (ASR) system to support multiple languages is challenging because the sub-word unit, lexicon and word inventories are typically language specific. In contrast, sequence-to-sequence models are well suited for multilingual ASR because they encapsulate an acoustic, pronunciation and language model jointly in a single network. In this work we present a single sequence-to-sequence ASR model trained on 9 different Indian languages, which have very little overlap in their scripts. Specifically, we take a union of language-specific grapheme sets and train a grapheme-based sequence-to-sequence model jointly on data from all languages. We find that this model, which is not explicitly given any information about language identity, improves recognition performance by 21% relative compared to analogous sequence-to-sequence models trained on each language individually. By modifying the model to accept a language identifier as an additional input feature, we further improve performance by an additional 7% relative and eliminate confusion between different languages.

Unsupervised image-to-image translation aims at learning a joint distribution of images in different domains by using images from the marginal distributions in individual domains. Since there exists an infinite set of joint distributions that can arrive the given marginal distributions, one could infer nothing about the joint distribution from the marginal distributions without additional assumptions. To address the problem, we make a shared-latent space assumption and propose an unsupervised image-to-image translation framework based on Coupled GANs. We compare the proposed framework with competing approaches and present high quality image translation results on various challenging unsupervised image translation tasks, including street scene image translation, animal image translation, and face image translation. We also apply the proposed framework to domain adaptation and achieve state-of-the-art performance on benchmark datasets. Code and additional results are available in https://github.com/mingyuliutw/unit .

In my previous blog, I looked at how quickly Artificial Intelligence (AI) is rapidly becoming a part of the retail experience. Industry analysts agree that 2018 will be the year that AI brings top and bottom line benefits to innovative companies. So, where can AI most help retailers? Here's four questions I think a combination of AI and analytics are perfectly suited to answer. What do you think about being served by a robot?

The prediction of the gas production from mature gas wells, due to their complex end-of-life behavior, is challenging and crucial for operational decision making. In this paper, we apply a modified deep LSTM model for prediction of the gas flow rates in mature gas wells, including the uncertainties in input parameters. Additionally, due to changes in the system in time and in order to increase the accuracy and robustness of the prediction, the Ensemble Kalman Filter (EnKF) is used to update the flow rate predictions based on new observations. The developed approach was tested on the data from two mature gas production wells in which their production is highly dynamic and suffering from salt deposition. The results show that the flow predictions using the EnKF updated model leads to better Jeffreys' J-divergences than the predictions without the EnKF model updating scheme.

Patients in the intensive care unit (ICU) require constant and close supervision. To assist clinical staff in this task, hospitals use monitoring systems that trigger audiovisual alarms if their algorithms indicate that a patient's condition may be worsening. However, current monitoring systems are extremely sensitive to movement artefacts and technical errors. As a result, they typically trigger hundreds to thousands of false alarms per patient per day - drowning the important alarms in noise and adding to the exhaustion of clinical staff. In this setting, data is abundantly available, but obtaining trustworthy annotations by experts is laborious and expensive. We frame the problem of false alarm reduction from multivariate time series as a machine-learning task and address it with a novel multitask network architecture that utilises distant supervision through multiple related auxiliary tasks in order to reduce the number of expensive labels required for training. We show that our approach leads to significant improvements over several state-of-the-art baselines on real-world ICU data and provide new insights on the importance of task selection and architectural choices in distantly supervised multitask learning.