Preference Neural Network (PNN)
This paper proposes a preference neural network (PNN) to address the problem of indifference preferences orders with new activation function. PNN also solves the Multi-label ranking problem, where labels may have indifference preference orders or subgroups are equally ranked. PNN follows a multi-layer feedforward architecture with fully connected neurons. Each neuron contains a novel smooth stairstep activation function based on the number of preference orders. PNN inputs represent data features and output neurons represent label indexes. The proposed PNN is evaluated using new preference mining dataset that contains repeated label values which have not experimented before. PNN outperforms five previously proposed methods for strict label ranking in terms of accurate results with high computational efficiency. …

Guided Complement Entropy (GCE)
Model robustness has been an important issue, since adding small adversarial perturbations to images is sufficient to drive the model accuracy down to nearly zero. In this paper, we propose a new training objective ‘Guided Complement Entropy’ (GCE) that has dual desirable effects: (a) neutralizing the predicted probabilities of incorrect classes, and (b) maximizing the predicted probability of the ground-truth class, particularly when (a) is achieved. Training with GCE encourages models to learn latent representations where samples of different classes form distinct clusters, which we argue, improves the model robustness against adversarial perturbations. Furthermore, compared with the state-of-the-arts trained with cross-entropy, same models trained with GCE achieve significant improvements on the robustness against white-box adversarial attacks, both with and without adversarial training. When no attack is present, training with GCE also outperforms cross-entropy in terms of model accuracy. …

WaveGlow
In this paper we propose WaveGlow: a flow-based network capable of generating high quality speech from mel-spectrograms. WaveGlow combines insights from Glow and WaveNet in order to provide fast, efficient and high-quality audio synthesis, without the need for auto-regression. WaveGlow is implemented using only a single network, trained using only a single cost function: maximizing the likelihood of the training data, which makes the training procedure simple and stable. Our PyTorch implementation produces audio samples at a rate of more than 500 kHz on an NVIDIA V100 GPU. Mean Opinion Scores show that it delivers audio quality as good as the best publicly available WaveNet implementation. All code will be made publicly available online. …

Weakly Structured Information Processing and Exploration (WIPE)
WIPE is used for managing the graph traversal manipulation with BI-like data aggregation. WIPE stands for “Weakly-structured Information Processing and Exploration”. It is a data manipulation and query language built on top of the graph functionality in the SAP HANA Database. Like other domain specific languages provided by SAP HANA Database, WIPE is embedded in transactional context, which means that multiple WIPE statements can be executed concurrently, guaranteeing the atomicity, consistency, isolation and durability. With the help of this language, multiple graph operations such as inserting, updating or deleting a node and other query operations can be declared in one complex statement. It is the graph abstraction layer in the SAP HANA Database that provides interaction with the graph data stored in the database by exposing graph concepts directly to the application developer. The application developer can create or delete graphs, access the existing graphs, modify the vertices and edges of the graphs, or retrieve a set of vertices and edges based on their attributes. Besides retrieval and manipulation functions, a set of built-in graph operators are also provided by the SAP HANA Database. These operators, such as breadth-first or depth-first traversal algorithms, interact with the column store of the relational engine to execute efficiently and in a highly optimum manner. …

AR-Annotator
An increasing number of scientific publications are created in open and transparent peer review models: a submission is published first, and then reviewers are invited, or a submission is reviewed in a closed environment but then these reviews are published with the final article, or combinations of these. Reasons for open peer review include giving better credit to reviewers and enabling readers to better appraise the quality of a publication. In most cases, the full, unstructured text of an open review is published next to the full, unstructured text of the article reviewed. This approach prevents human readers from getting a quick impression of the quality of parts of an article, and it does not easily support secondary exploitation, e.g., for scientometrics on reviews. While document formats have been proposed for publishing structured articles including reviews, integrated tool support for entire open peer review workflows resulting in such documents is still scarce. We present AR-Annotator, the Automatic Article and Review Annotator which employs a semantic information model of an article and its reviews, using semantic markup and unique identifiers for all entities of interest. The fine-grained article structure is not only exposed to authors and reviewers but also preserved in the published version. We publish articles and their reviews in a Linked Data representation and thus maximize their reusability by third-party applications. We demonstrate this reusability by running quality-related queries against the structured representation of articles and their reviews. …

AgentGraph
Dialogue policy plays an important role in task-oriented spoken dialogue systems. It determines how to respond to users. The recently proposed deep reinforcement learning (DRL) approaches have been used for policy optimization. However, these deep models are still challenging for two reasons: 1) Many DRL-based policies are not sample-efficient. 2) Most models don’t have the capability of policy transfer between different domains. In this paper, we propose a universal framework, AgentGraph, to tackle these two problems. The proposed AgentGraph is the combination of GNN-based architecture and DRL-based algorithm. It can be regarded as one of the multi-agent reinforcement learning approaches. Each agent corresponds to a node in a graph, which is defined according to the dialogue domain ontology. When making a decision, each agent can communicate with its neighbors on the graph. Under AgentGraph framework, we further propose Dual GNN-based dialogue policy, which implicitly decomposes the decision in each turn into a high-level global decision and a low-level local decision. Experiments show that AgentGraph models significantly outperform traditional reinforcement learning approaches on most of the 18 tasks of the PyDial benchmark. Moreover, when transferred from the source task to a target task, these models not only have acceptable initial performance but also converge much faster on the target task. …

Latent Class-Conditional Noise Model (LCCN)
Learning with noisy labels, which aims to reduce expensive labors on accurate annotations, has become imperative in the Big Data era. Previous noise transition based method has achieved promising results and presented a theoretical guarantee on performance in the case of class-conditional noise. However, this type of approaches critically depend on an accurate pre-estimation of the noise transition, which is usually impractical. Subsequent improvement adapts the pre-estimation along with the training progress via a Softmax layer. However, the parameters in the Softmax layer are highly tweaked for the fragile performance due to the ill-posed stochastic approximation. To address these issues, we propose a Latent Class-Conditional Noise model (LCCN) that naturally embeds the noise transition under a Bayesian framework. By projecting the noise transition into a Dirichlet-distributed space, the learning is constrained on a simplex based on the whole dataset, instead of some ad-hoc parametric space. We then deduce a dynamic label regression method for LCCN to iteratively infer the latent labels, to stochastically train the classifier and to model the noise. Our approach safeguards the bounded update of the noise transition, which avoids previous arbitrarily tuning via a batch of samples. We further generalize LCCN for open-set noisy labels and the semi-supervised setting. We perform extensive experiments with the controllable noise data sets, CIFAR-10 and CIFAR-100, and the agnostic noise data sets, Clothing1M and WebVision17. The experimental results have demonstrated that the proposed model outperforms several state-of-the-art methods. …

Image Registration
Image registration is the process of transforming different sets of data into one coordinate system. Data may be multiple photographs, data from different sensors, times, depths, or viewpoints. It is used in computer vision, medical imaging, military automatic target recognition, and compiling and analyzing images and data from satellites. Registration is necessary in order to be able to compare or integrate the data obtained from these different measurements. …

Fenwick Tree
A Fenwick tree or binary indexed tree is a data structure that can efficiently update elements and calculate prefix sums in a table of numbers. This structure was proposed by Peter Fenwick in 1994 to improve the efficiency of arithmetic coding compression algorithms. When compared with a flat array of numbers, the Fenwick tree achieves a much better balance between two operations: element update and prefix sum calculation. In a flat array of {\displaystyle n} n numbers, you can either store the elements, or the prefix sums. In the first case, computing prefix sums requires linear time; in the second case, updating the array elements requires linear time (in both cases, the other operation can be performed in constant time). Fenwick trees allow both operations to be performed in {\displaystyle O(\log n)} O(\log n) time. This is achieved by representing the numbers as a tree, where the value of each node is the sum of the numbers in that subtree. The tree structure allows operations to be performed using only {\displaystyle O(\log n)} O(\log n) node accesses. …

Mechanical Turk (MTurk)
Amazon Mechanical Turk (MTurk) is a crowdsourcing Internet marketplace that enables individuals and businesses (known as Requesters) to coordinate the use of human intelligence to perform tasks that computers are currently unable to do. It is one of the sites of Amazon Web Services. Employers are able to post jobs known as HITs (Human Intelligence Tasks), such as choosing the best among several photographs of a storefront, writing product descriptions, or identifying performers on music CDs. Workers (called Providers in Mechanical Turk’s Terms of Service, or, more colloquially, Turkers) can then browse among existing jobs and complete them for a monetary payment set by the employer. To place jobs, the requesting programs use an open application programming interface (API), or the more limited MTurk Requester site. Employers are restricted to US-based entities. …

Supervised Quantile Normalisation (SUQUAN)
Quantile normalisation is a popular normalisation method for data subject to unwanted variations such as images, speech, or genomic data. It applies a monotonic transformation to the feature values of each sample to ensure that after normalisation, they follow the same target distribution for each sample. Choosing a ‘good’ target distribution remains however largely empirical and heuristic, and is usually done independently of the subsequent analysis of normalised data. We propose instead to couple the quantile normalisation step with the subsequent analysis, and to optimise the target distribution jointly with the other parameters in the analysis. We illustrate this principle on the problem of estimating a linear model over normalised data, and show that it leads to a particular low-rank matrix regression problem that can be solved efficiently. We illustrate the potential of our method, which we term SUQUAN, on simulated data, images and genomic data, where it outperforms standard quantile normalisation. …

DeepFool
State-of-the-art deep neural networks have achieved impressive results on many image classification tasks. However, these same architectures have been shown to be unstable to small, well sought, perturbations of the images. Despite the importance of this phenomenon, no effective methods have been proposed to accurately compute the robustness of state-of-the-art deep classifiers to such perturbations on large-scale datasets. In this paper, we fill this gap and propose the DeepFool algorithm to efficiently compute perturbations that fool deep networks, and thus reliably quantify the robustness of these classifiers. Extensive experimental results show that our approach outperforms recent methods in the task of computing adversarial perturbations and making classifiers more robust
DeepFool – A simple and accurate method to fool Deep Neural Networks. …

MCES-P
Optimal decision making with limited or no information in stochastic environments where multiple agents interact is a challenging topic in the realm of artificial intelligence. Reinforcement learning (RL) is a popular approach for arriving at optimal strategies by predicating stimuli, such as the reward for following a strategy, on experience. RL is heavily explored in the single-agent context, but is a nascent concept in multiagent problems. To this end, I propose several principled model-free and partially model-based reinforcement learning approaches for several multiagent settings. In the realm of normative reinforcement learning, I introduce scalable extensions to Monte Carlo exploring starts for partially observable Markov Decision Processes (POMDP), dubbed MCES-P, where I expand the theory and algorithm to the multiagent setting. I first examine MCES-P with probably approximately correct (PAC) bounds in the context of multiagent setting, showing MCESP+PAC holds in the presence of other agents. I then propose a more sample-efficient methodology for antagonistic settings, MCESIP+PAC. For cooperative settings, I extend MCES-P to the Multiagent POMDP, dubbed MCESMP+PAC. I then explore the use of reinforcement learning as a methodology in searching for optima in realistic and latent model environments. First, I explore a parameterized Q-learning approach in modeling humans learning to reason in an uncertain, multiagent environment. Next, I propose an implementation of MCES-P, along with image segmentation, to create an adaptive team-based reinforcement learning technique to positively identify the presence of phenotypically-expressed water and pathogen stress in crop fields. …

ConRelMiner
The General Data Protection Regulation (GDPR) poses enormous challenges on companies and organizations with respect to understanding, implementing, and maintaining the contained constraints. We report on how the ConRelMiner method can be used for untangling the GDPR. For this, the GDPR is filtered and grouped along the roles mentioned by the GDPR and the reduction of sentences to be read by analysts is shown. Moreover, the output of the ConRelMiner – a cluster graph with relations between the sentences – is displayed and interpreted. Overall the goal is to illustrate how the effort for implementing the GDPR can be reduced and a structured and meaningful representation of the relevant GDPR sentences can be found. …

Datar
Various tools, softwares and systems are proposed and implemented to tackle the challenges in big data on different emphases, e.g., data analysis, data transaction, data query, data storage, data visualization, data privacy. In this paper, we propose datar, a new prospective and unified framework for Big Data Management System (BDMS) from the point of system architecture by leveraging ideas from mainstream computer structure. We introduce five key components of datar by reviewing the current status of BDMS. Datar features with configuration chain of pluggable engines, automatic dataflow on job pipelines, intelligent self-driving system management and interactive user interfaces. Moreover, we present biggy as an implementation of datar with manipulation details demonstrated by four running examples. Evaluations on efficiency and scalability are carried out to show the performance. Our work argues that the envisioned datar is a feasible solution to the unified framework of BDMS, which can manage big data pluggablly, automatically and intelligently with specific functionalities, where specific functionalities refer to input, storage, computation, control and output of big data. …

SupportNet
A plain well-trained deep learning model often does not have the ability to learn new knowledge without forgetting the previously learned knowledge, which is known as the catastrophic forgetting. Here we propose a novel method, SupportNet, to solve the catastrophic forgetting problem in class incremental learning scenario efficiently and effectively. SupportNet combines the strength of deep learning and support vector machine (SVM), where SVM is used to identify the support data from the old data, which are fed to the deep learning model together with the new data for further training so that the model can review the essential information of the old data when learning the new information. Two powerful consolidation regularizers are applied to ensure the robustness of the learned model. Comprehensive experiments on various tasks, including enzyme function prediction, subcellular structure classification and breast tumor classification, show that SupportNet drastically outperforms the state-of-the-art incremental learning methods and even reaches similar performance as the deep learning model trained from scratch on both old and new data. Our program is accessible at: https://…/SupportNet …

GPyTorch
Despite advances in scalable models, the inference tools used for Gaussian processes (GPs) have yet to fully capitalize on recent developments in machine learning hardware. We present an efficient and general approach to GP inference based on Blackbox Matrix-Matrix multiplication (BBMM). BBMM inference uses a modified batched version of the conjugate gradients algorithm to derive all terms required for training and inference in a single call. Adapting this algorithm to complex models simply requires a routine for efficient matrix-matrix multiplication with the kernel and its derivative. In addition, BBMM utilizes a specialized preconditioner that substantially speeds up convergence. In experiments, we show that BBMM efficiently utilizes GPU hardware, speeding up GP inference by an order of magnitude on a variety of popular GP models. Additionally, we provide GPyTorch, a new software platform for scalable Gaussian process inference via BBMM, built on PyTorch. …

Modified Q-Learner for the Vasicek Model (MQLV)
In a reinforcement learning approach, an optimal value function is learned across a set of actions, or decisions, that leads to a set of states giving different rewards, with the objective to maximize the overall reward. A policy assigns to each state-action pairs an expected return. We call an optimal policy a policy for which the value function is optimal. QLBS, Q-Learner in the Black-Scholes(-Merton) Worlds, applies the reinforcement learning concepts, and noticeably, the popular Q-learning algorithm, to the financial stochastic model described by Black, Scholes and Merton. However, QLBS is specifically optimized for the geometric Brownian motion and the pricing of vanilla options. Consequently, it suffers from the traditional over-estimation of the Q-values reflected by an over-estimation of the vanilla option prices. Furthermore, its range of application is limited to vanilla option pricing within the financial markets. We propose MQLV, Modified Q-Learner for the Vasicek model, a new reinforcement learning approach that limits the Q-values over-estimation observed in QLBS and extends the simulation to mean reverting stochastic diffusion processes. Additionally, MQLV uses a digital function to estimate the future probability of an event, thus widening the scope of the financial application to any other domain involving time series. Our experiments underline the potential of MQLV on generated Monte Carlo simulations, particularly representative of the retail banking time series. In particular, MQLV is able to determine the optimal policy of money management based on the aggregated financial transactions of the clients, unlocking new frontiers to establish personalized credit card limits or loans. Finally, MQLV is the first methodology compatible with the Vasicek model capable of an event probability estimation targeting simulation of event probabilities in retail banking. …

ease.ml/ci
Continuous integration is an indispensable step of modern software engineering practices to systematically manage the life cycles of system development. Developing a machine learning model is no difference – it is an engineering process with a life cycle, including design, implementation, tuning, testing, and deployment. However, most, if not all, existing continuous integration engines do not support machine learning as first-class citizens. In this paper, we present ease.ml/ci, to our best knowledge, the first continuous integration system for machine learning. The challenge of building ease.ml/ci is to provide rigorous guarantees, e.g., single accuracy point error tolerance with 0.999 reliability, with a practical amount of labeling effort, e.g., 2K labels per test. We design a domain specific language that allows users to specify integration conditions with reliability constraints, and develop simple novel optimizations that can lower the number of labels required by up to two orders of magnitude for test conditions popularly used in real production systems. …

AgrLearn
We establish an equivalence between information bottleneck (IB) learning and an unconventional quantization problem, `IB quantization’. Under this equivalence, standard neural network models correspond to scalar IB quantizers. We prove a coding theorem for IB quantization, which implies that scalar IB quantizers are in general inferior to vector IB quantizers. This inspires us to develop a learning framework for neural networks, AgrLearn, that corresponds to vector IB quantizers. We experimentally verify that AgrLearn applied to some deep network models of current art improves upon them, while requiring less training data. With a heuristic smoothing, AgrLearn further improves its performance, resulting in new state of the art in image classification on Cifar10. …

Disaggregation
Disaggregation is the breakdown of observations, usually within a common branch of a hierarchy, to a more detailed level to that at which detailed observations are taken. …

Deeply Supervised Object Detector (DSOD)
We propose Deeply Supervised Object Detectors (DSOD), an object detection framework that can be trained from scratch. Recent advances in object detection heavily depend on the off-the-shelf models pre-trained on large-scale classification datasets like ImageNet and OpenImage. However, one problem is that adopting pre-trained models from classification to detection task may incur learning bias due to the different objective function and diverse distributions of object categories. Techniques like fine-tuning on detection task could alleviate this issue to some extent but are still not fundamental. Furthermore, transferring these pre-trained models across discrepant domains will be more difficult (e.g., from RGB to depth images). Thus, a better solution to handle these critical problems is to train object detectors from scratch, which motivates our proposed method. Previous efforts on this direction mainly failed by reasons of the limited training data and naive backbone network structures for object detection. In DSOD, we contribute a set of design principles for learning object detectors from scratch. One of the key principles is the deep supervision, enabled by layer-wise dense connections in both backbone networks and prediction layers, plays a critical role in learning good detectors from scratch. After involving several other principles, we build our DSOD based on the single-shot detection framework (SSD). We evaluate our method on PASCAL VOC 2007, 2012 and COCO datasets. DSOD achieves consistently better results than the state-of-the-art methods with much more compact models. Specifically, DSOD outperforms baseline method SSD on all three benchmarks, while requiring only 1/2 parameters. We also observe that DSOD can achieve comparable/slightly better results than Mask RCNN + FPN (under similar input size) with only 1/3 parameters, using no extra data or pre-trained models. …

PAI Data
The Project PAI Data Protocol (‘PAI Data’) is a specification that extends the Project PAI Blockchain Protocol to include a method of securing and provisioning access to arbitrary data. In the context of PAI Coin Development Proposal (PDP) 2, this paper defines two important transaction types that PAI Data supports: Storage Transactions, which facilitate storage of data and proof of ownership, and Sharing Transactions, designed to enable granting and revocation of data access to designated recipients. A comparative analysis of PAI Data against similar blockchain-based file storage systems is also presented. …

Average Shifted Histogram (ASH)
A simple device has been proposed for eliminating the bin edge problem of the frequency polygon while retaining many of the computational advantages of a density estimate based on bin counts. Scott (1983, 1985b) considered the problem of choosing among the collection of multivariate frequency polygons, each with the same smoothing parameter but di ering bin origins. Rather than choosing the \smoothest’ such curve or surface, he proposed averaging several of the shifted frequency polygons. As the average of piecewise linear curves is also piecewise linear, the resulting curve appears to be a frequency polygon as well. If the weights are nonnegative and sum to 1, the resulting \averaged shifted frequency polygon’ (ASFP) is nonnegative and integrates to 1. A nearly equivalent device is to average several shifted histograms, which is just as general but simpler to describe and analyze. The result is the \averaged shifted histogram’ (ASH). Since the average of piecewise constant functions such as the histogram is also piecewise constant, the ASH appears to be a histogram as well. In practice, the ASH is made continuous using either of the linear interpolation schemes described for the frequency polygon in Chapter 4 and will be referred to as the frequency polygon of the averaged shifted histogram (FP-ASH). The ASH is the practical choice for computationally and statistically e cient density estimation. …

Metalog Distribution
In economics, business, engineering, science and other fields, continuous uncertainties frequently arise that are not easily- or well-characterized by previously-named continuous probability distributions. Frequently, there is data available from measurements, assessments, derivations, simulations or other sources that characterize the range of an uncertainty. But the underlying process that generated this data is either unknown or fails to lend itself to convenient derivation of equations that appropriately characterize the probability density (PDF), cumulative (CDF) or quantile distribution functions. The metalog distributions are a family of continuous univariate probability distributions that directly address this need. They can be used in most any situation in which CDF data is known and a flexible, simple, and easy-to-use continuous probability distribution is needed to represent that data. Consider their uses and benefits. Also consider their applications over a wide range of fields and data sources. …

Long Short-Term Attention (LSTA)
Egocentric activity recognition is one of the most challenging tasks in video analysis. It requires a fine-grained discrimination of small objects and their manipulation. While some methods base on strong supervision and attention mechanisms, they are either annotation consuming or do not take spatio-temporal patterns into account. In this paper we propose LSTA as a mechanism to focus on features from spatial relevant parts while attention is being tracked smoothly across the video sequence. We demonstrate the effectiveness of LSTA on egocentric activity recognition with an end-to-end trainable two-stream architecture, achieving state of the art performance on four standard benchmarks. …

Gaussian Process Posterior Sampling Reinforcement Learning (GPPSTD)
Efficient Reinforcement Learning usually takes advantage of demonstration or good exploration strategy. By applying posterior sampling in model-free RL under the hypothesis of GP, we propose Gaussian Process Posterior Sampling Reinforcement Learning(GPPSTD) algorithm in continuous state space, giving theoretical justifications and empirical results. We also provide theoretical and empirical results that various demonstration could lower expected uncertainty and benefit posterior sampling exploration. In this way, we combined the demonstration and exploration process together to achieve a more efficient reinforcement learning. …

SkinnerDB
SkinnerDB is designed from the ground up for reliable join ordering. It maintains no data statistics and uses no cost or cardinality models. Instead, it uses reinforcement learning to learn optimal join orders on the fly, during the execution of the current query. To that purpose, we divide the execution of a query into many small time slices. Different join orders are tried in different time slices. We merge result tuples generated according to different join orders until a complete result is obtained. By measuring execution progress per time slice, we identify promising join orders as execution proceeds. Along with SkinnerDB, we introduce a new quality criterion for query execution strategies. We compare expected execution cost against execution cost for an optimal join order. SkinnerDB features multiple execution strategies that are optimized for that criterion. Some of them can be executed on top of existing database systems. For maximal performance, we introduce a customized execution engine, facilitating fast join order switching via specialized multi-way join algorithms and tuple representations. We experimentally compare SkinnerDB’s performance against various baselines, including MonetDB, Postgres, and adaptive processing methods. We consider various benchmarks, including the join order benchmark and TPC-H variants with user-defined functions. Overall, the overheads of reliable join ordering are negligible compared to the performance impact of the occasional, catastrophic join order choice. …

Cubic Long Short-Term Memory (CubicLSTM)
Predicting future frames in videos has become a promising direction of research for both computer vision and robot learning communities. The core of this problem involves moving object capture and future motion prediction. While object capture specifies which objects are moving in videos, motion prediction describes their future dynamics. Motivated by this analysis, we propose a Cubic Long Short-Term Memory (CubicLSTM) unit for video prediction. CubicLSTM consists of three branches, i.e., a spatial branch for capturing moving objects, a temporal branch for processing motions, and an output branch for combining the first two branches to generate predicted frames. Stacking multiple CubicLSTM units along the spatial branch and output branch, and then evolving along the temporal branch can form a cubic recurrent neural network (CubicRNN). Experiment shows that CubicRNN produces more accurate video predictions than prior methods on both synthetic and real-world datasets. …

Neumann Optimizer
Progress in deep learning is slowed by the days or weeks it takes to train large models. The natural solution of using more hardware is limited by diminishing returns, and leads to inefficient use of additional resources. In this paper, we present a large batch, stochastic optimization algorithm that is both faster than widely used algorithms for fixed amounts of computation, and also scales up substantially better as more computational resources become available. Our algorithm implicitly computes the inverse Hessian of each mini-batch to produce descent directions; we do so without either an explicit approximation to the Hessian or Hessian-vector products. We demonstrate the effectiveness of our algorithm by successfully training large ImageNet models (Inception-V3, Resnet-50, Resnet-101 and Inception-Resnet-V2) with mini-batch sizes of up to 32000 with no loss in validation error relative to current baselines, and no increase in the total number of steps. At smaller mini-batch sizes, our optimizer improves the validation error in these models by 0.8-0.9%. Alternatively, we can trade off this accuracy to reduce the number of training steps needed by roughly 10-30%. Our work is practical and easily usable by others — only one hyperparameter (learning rate) needs tuning, and furthermore, the algorithm is as computationally cheap as the commonly used Adam optimizer. …

Self Other-Modeling (SOM)
We consider the multi-agent reinforcement learning setting with imperfect information in which each agent is trying to maximize its own utility. The reward function depends on the hidden state (or goal) of both agents, so the agents must infer the other players’ hidden goals from their observed behavior in order to solve the tasks. We propose a new approach for learning in these domains: Self Other-Modeling (SOM), in which an agent uses its own policy to predict the other agent’s actions and update its belief of their hidden state in an online manner. We evaluate this approach on three different tasks and show that the agents are able to learn better policies using their estimate of the other players’ hidden states, in both cooperative and adversarial settings. …

Statistically Discriminative Sub-trajectory Mining (SDSM)
We study the problem of discriminative sub-trajectory mining. Given two groups of trajectories, the goal of this problem is to extract moving patterns in the form of sub-trajectories which are more similar to sub-trajectories of one group and less similar to those of the other. We propose a new method called Statistically Discriminative Sub-trajectory Mining (SDSM) for this problem. An advantage of the SDSM method is that the statistical significance of the extracted sub-trajectories are properly controlled in the sense that the probability of finding a false positive sub-trajectory is smaller than a specified significance threshold alpha (e.g., 0.05), which is indispensable when the method is used in scientific or social studies under noisy environment. Finding such statistically discriminative sub-trajectories from massive trajectory dataset is both computationally and statistically challenging. In the SDSM method, we resolve the difficulties by introducing a tree representation among sub-trajectories and running an efficient permutation-based statistical inference method on the tree. To the best of our knowledge, SDSM is the first method that can efficiently extract statistically discriminative sub-trajectories from massive trajectory dataset. We illustrate the effectiveness and scalability of the SDSM method by applying it to a real-world dataset with 1,000,000 trajectories which contains 16,723,602,505 sub-trajectories. …

Apache Forrest
Apache Forrest software is a publishing framework that transforms input from various sources into a unified presentation in one or more output formats. The modular and extensible plug-in architecture of Apache Forrest is based on Apache Cocoon and the relevant industry standards that separate presentation from content. Forrest can generate static documents, or be used as a dynamic server, or be deployed by its automated facility. …

Internet of NanoThing (IoNT)
This chapter focuses on Internet of Things from the nanoscale point of view. The chapter starts with section 1 which provides an introduction of nanothings and nanotechnologies. The nanoscale communication paradigms and the different approaches are discussed for nanodevices development. Nanodevice characteristics are discussed and the architecture of wireless nanodevices are outlined. Section 2 describes Internet of NanoThing(IoNT), its network architecture, and the challenges of nanoscale communication which is essential for enabling IoNT. Section 3 gives some practical applications of IoNT. The internet of Bio-NanoThing (IoBNT) and relevant biomedical applications are discussed. Other Applications such as military, industrial, and environmental applications are also outlined. …

Neural Architecture Optimization (NAO)
Automatic neural architecture design has shown its potential in discovering powerful neural network architectures. Existing methods, no matter based on reinforcement learning or evolutionary algorithms (EA), conduct architecture search in a discrete space, which is highly inefficient. In this paper, we propose a simple and efficient method to automatic neural architecture design based on continuous optimization. We call this new approach neural architecture optimization (NAO). There are three key components in our proposed approach: (1) An encoder embeds/maps neural network architectures into a continuous space. (2) A predictor takes the continuous representation of a network as input and predicts its accuracy. (3) A decoder maps a continuous representation of a network back to its architecture. The performance predictor and the encoder enable us to perform gradient based optimization in the continuous space to find the embedding of a new architecture with potentially better accuracy. Such a better embedding is then decoded to a network by the decoder. Experiments show that the architecture discovered by our method is very competitive for image classification task on CIFAR-10 and language modeling task on PTB, outperforming or on par with the best results of previous architecture search methods with a significantly reduction of computational resources. Specifically we obtain $2.07\%$ test set error rate for CIFAR-10 image classification task and $55.9$ test set perplexity of PTB language modeling task. The best discovered architectures on both tasks are successfully transferred to other tasks such as CIFAR-100 and WikiText-2. …

Active Learning with Partial Feedback
In the large-scale multiclass setting, assigning labels often consists of answering multiple questions to drill down through a hierarchy of classes. Here, the labor required per annotation scales with the number of questions asked. We propose active learning with partial feedback. In this setup, the learner asks the annotator if a chosen example belongs to a (possibly composite) chosen class. The answer eliminates some classes, leaving the agent with a partial label. Success requires (i) a sampling strategy to choose (example, class) pairs, and (ii) learning from partial labels. Experiments on the TinyImageNet dataset demonstrate that our most effective method achieves a 21% relative improvement in accuracy for a 200k binary question budget. Experiments on the TinyImageNet dataset demonstrate that our most effective method achieves a 26% relative improvement (8.1% absolute) in top1 classification accuracy for a 250k (or 30%) binary question budget, compared to a naive baseline. Our work may also impact traditional data annotation. For example, our best method fully annotates TinyImageNet with only 482k (with EDC though, ERC is 491) binary questions (vs 827k for naive method). …

Regularized Contextual Bandits
We consider the stochastic contextual bandit problem with additional regularization. The motivation comes from problems where the policy of the agent must be close to some baseline policy which is known to perform well on the task. To tackle this problem we use a nonparametric model and propose an algorithm splitting the context space into bins, and solving simultaneously – and independently – regularized multi-armed bandit instances on each bin. We derive slow and fast rates of convergence, depending on the unknown complexity of the problem. We also consider a new relevant margin condition to get problem-independent convergence rates, ending up in intermediate convergence rates interpolating between the aforementioned slow and fast rates. …

Mother Compact Recurrent Memory (MCRM)
LSTMs and GRUs are the most common recurrent neural network architectures used to solve temporal sequence problems. The two architectures have differing data flows dealing with a common component called the cell state also referred to as the memory. We attempt to enhance the memory by presenting a biologically inspired modification that we call the Mother Compact Recurrent Memory MCRM. MCRMs are a type of a nested LSTM-GRU architecture where the cell state is the GRU’s hidden state. The relationship between the womb and the fetus is analogous to the relationship between the LSTM and GRU inside MCRM in that the fetus is connected to its womb through the umbilical cord. The umbilical cord consists of two arteries and one vein. The two arteries are considered as an input to the fetus which is analogous to the concatenation of the forget gate and input gate from the LSTM. The vein is the output from the fetus which plays the role of the hidden state of the GRU. Because MCRMs has this type of nesting, MCRMs have a compact memory pattern consisting of neurons that act explicitly in both long-term and short-term fashions. For some specific tasks, empirical results show that MCRMs outperform previously used architectures. …

Neuromorphic Engineering
Neuromorphic engineering, also known as neuromorphic computing, is a concept developed by Carver Mead, in the late 1980s, describing the use of very-large-scale integration (VLSI) systems containing electronic analog circuits to mimic neuro-biological architectures present in the nervous system. In recent times, the term neuromorphic has been used to describe analog, digital, mixed-mode analog/digital VLSI, and software systems that implement models of neural systems (for perception, motor control, or multisensory integration). The implementation of neuromorphic computing on the hardware level can be realized by oxide-based memristors,, spintronic memories, threshold switches, and transistors. A key aspect of neuromorphic engineering is understanding how the morphology of individual neurons, circuits, applications, and overall architectures creates desirable computations, affects how information is represented, influences robustness to damage, incorporates learning and development, adapts to local change (plasticity), and facilitates evolutionary change. Neuromorphic engineering is an interdisciplinary subject that takes inspiration from biology, physics, mathematics, computer science, and electronic engineering to design artificial neural systems, such as vision systems, head-eye systems, auditory processors, and autonomous robots, whose physical architecture and design principles are based on those of biological nervous systems. …

Neural Network based Collaborative Ranking (NCR)
Recommender systems are aimed at generating a personalized ranked list of items that an end user might be interested in. With the unprecedented success of deep learning in computer vision and speech recognition, recently it has been a hot topic to bridge the gap between recommender systems and deep neural network. And deep learning methods have been shown to achieve state-of-the-art on many recommendation tasks. For example, a recent model, NeuMF, first projects users and items into some shared low-dimensional latent feature space, and then employs neural nets to model the interaction between the user and item latent features to obtain state-of-the-art performance on the recommendation tasks. NeuMF assumes that the non-interacted items are inherent negative and uses negative sampling to relax this assumption. In this paper, we examine an alternative approach which does not assume that the non-interacted items are necessarily negative, just that they are less preferred than interacted items. Specifically, we develop a new classification strategy based on the widely used pairwise ranking assumption. We combine our classification strategy with the recently proposed neural collaborative filtering framework, and propose a general collaborative ranking framework called Neural Network based Collaborative Ranking (NCR). We resort to a neural network architecture to model a user’s pairwise preference between items, with the belief that neural network will effectively capture the latent structure of latent factors. The experimental results on two real-world datasets show the superior performance of our models in comparison with several state-of-the-art approaches. …