Binci
Binci is a utility that allows you to easily containerize your development workflow using Docker. Simply put, it’s like having a cleanroom for all of your development processes which contain services (like databases) without needing to setup and maintain these environments manually. …

CryptoNN
Emerging neural networks based machine learning techniques such as deep learning and its variants have shown tremendous potential in many application domains. However, they raise serious privacy concerns due to the risk of leakage of highly privacy-sensitive data when data collected from users is used to train neural network models to support predictive tasks. To tackle such serious privacy concerns, several privacy-preserving approaches have been proposed in the literature that use either secure multi-party computation (SMC) or homomorphic encryption (HE) as the underlying mechanisms. However, neither of these cryptographic approaches provides an efficient solution towards constructing a privacy-preserving machine learning model, as well as supporting both the training and inference phases. To tackle the above issue, we propose a CryptoNN framework that supports training a neural network model over encrypted data by using the emerging functional encryption scheme instead of SMC or HE. We also construct a functional encryption scheme for basic arithmetic computation to support the requirement of the proposed CryptoNN framework. We present performance evaluation and security analysis of the underlying crypto scheme and show through our experiments that CryptoNN achieves accuracy that is similar to those of the baseline neural network models on the MNIST dataset. …

Real-Time Predictive Analytics
It is when a predictive model (built/fitted on a set of aggregated data) is deployed to perform run-time prediction on a continuous stream of event data to enable decision making in real-time. In order to achieve this, there are two aspects involved. One, the predictive model built by a Data Scientist via a stand-alone tool (R, SAS, SPSS, etc.) has to be exported in a consumable format (PMML is a preferred method across machine learning environments these days; we have done this and also via other formats). Second, a streaming operational analytics platform has to consume the model (PMML or other format) and translate it into the necessary predictive function (via open-source jPMML or Cascading Pattern or Zementis’ commercial licensed UPPI or other interfaces), and also feed the processed streaming event data (via a stream processing component in CEP or similar) to compute the predicted outcome. This deployment of a complex predictive model, from its parent machine learning environment to an operational analytics environment, is one possible route in order to successfully achieve a continuous run-time prediction on streaming event data in real-time. …

KnowBias
We introduce KnowBias, a system for detecting the degree of political bias in textual content such as social media posts and news articles. In the space of scalable text classification, a common problem is domain mismatch, where easily accessible training data (i.e., tweets) does not correspond in format to the desired testing domain (i.e., longer form article content). While universal text encoders such as word or sentence embeddings could be leveraged to train target agnostic classifiers, such schemes result in poor performance on long-form articles. Our key insight is that long-form articles are a mix of neutral and political sentences, while tweets are concentrated with opinion. We propose a two-step classification system that first automatically filters out neutral sentences from the input text document at evaluation time, and then the resulting text is input into a polarity classifier. We evaluate our two-step approach using a variety of test suites, including a set of tweets and long-form articles where annotations were crowd-sourced to decrease label noise, measuring accuracy and Spearman-rho rank correlation. In practice, KnowBias achieves a high accuracy of 86% (rho = 0.65) on these tweets and 75% (rho = 0.69) on long-form articles. …

Denoising based Saliency Prediction with Generative Adversarial Network (DSAL-GAN)
Synthesizing high quality saliency maps from noisy images is a challenging problem in computer vision and has many practical applications. Samples generated by existing techniques for saliency detection cannot handle the noise perturbations smoothly and fail to delineate the salient objects present in the given scene. In this paper, we present a novel end-to-end coupled Denoising based Saliency Prediction with Generative Adversarial Network (DSAL-GAN) framework to address the problem of salient object detection in noisy images. DSAL-GAN consists of two generative adversarial-networks (GAN) trained end-to-end to perform denoising and saliency prediction altogether in a holistic manner. The first GAN consists of a generator which denoises the noisy input image, and in the discriminator counterpart we check whether the output is a denoised image or ground truth original image. The second GAN predicts the saliency maps from raw pixels of the input denoised image using a data-driven metric based on saliency prediction method with adversarial loss. Cycle consistency loss is also incorporated to further improve salient region prediction. We demonstrate with comprehensive evaluation that the proposed framework outperforms several baseline saliency models on various performance benchmarks. …

Graphonomy
Prior highly-tuned human parsing models tend to fit towards each dataset in a specific domain or with discrepant label granularity, and can hardly be adapted to other human parsing tasks without extensive re-training. In this paper, we aim to learn a single universal human parsing model that can tackle all kinds of human parsing needs by unifying label annotations from different domains or at various levels of granularity. This poses many fundamental learning challenges, e.g. discovering underlying semantic structures among different label granularity, performing proper transfer learning across different image domains, and identifying and utilizing label redundancies across related tasks. To address these challenges, we propose a new universal human parsing agent, named ‘Graphonomy’, which incorporates hierarchical graph transfer learning upon the conventional parsing network to encode the underlying label semantic structures and propagate relevant semantic information. In particular, Graphonomy first learns and propagates compact high-level graph representation among the labels within one dataset via Intra-Graph Reasoning, and then transfers semantic information across multiple datasets via Inter-Graph Transfer. Various graph transfer dependencies (\eg, similarity, linguistic knowledge) between different datasets are analyzed and encoded to enhance graph transfer capability. By distilling universal semantic graph representation to each specific task, Graphonomy is able to predict all levels of parsing labels in one system without piling up the complexity. Experimental results show Graphonomy effectively achieves the state-of-the-art results on three human parsing benchmarks as well as advantageous universal human parsing performance. …

ROWLTab
It has been argued that it is much easier to convey logical statements using rules rather than OWL (or description logic (DL)) axioms. Based on recent theoretical developments on transformations between rules and DLs, we have developed ROWLTab, a Protege plugin that allows users to enter OWL axioms by way of rules; the plugin then automatically converts these rules into OWL 2 DL axioms if possible, and prompts the user in case such a conversion is not possible without weakening the semantics of the rule. In this paper, we present ROWLTab, together with a user evaluation of its effectiveness compared to entering axioms using the standard Protege interface. Our evaluation shows that modeling with ROWLTab is much quicker than the standard interface, while at the same time, also less prone to errors for hard modeling tasks. …

HyperGAN
We introduce HyperGAN, a generative network that learns to generate all the weights within a deep neural network. HyperGAN employs a novel mixer to transform independent Gaussian noise into a latent space where dimensions are correlated, which is then transformed to generate weights in each layer of a deep neural network. We utilize an architecture that bears resemblance to generative adversarial networks, but we evaluate the likelihood of samples with a classification loss. This is equivalent to minimizing the KL-divergence between the generated network parameter distribution and an unknown true parameter distribution. We apply HyperGAN to classification, showing that HyperGAN can learn to generate parameters which solve the MNIST and CIFAR-10 datasets with competitive performance to fully supervised learning, while learning a rich distribution of effective parameters. We also show that HyperGAN can also provide better uncertainty than standard ensembles. This is evaluated by the ability of HyperGAN generated ensembles to detect out of distribution data as well as adversarial examples. We see that in addition to being highly accurate on inlier data, HyperGAN can provide reasonable uncertainty estimates. …

DLocRL
In recent years, with the prevalence of social media and smart devices, people causally reveal their locations such as shops, hotels, and restaurants in their tweets. Recognizing and linking such fine-grained location mentions to well-defined location profiles are beneficial for retrieval and recommendation systems. Prior studies heavily rely on hand-crafted linguistic features. Recently, deep learning has shown its effectiveness in feature representation learning for many NLP tasks. In this paper, we propose DLocRL, a new Deep pipeline for fine-grained Location Recognition and Linking in tweets. DLocRL leverages representation learning, semantic composition, attention and gate mechanisms to exploit semantic context features for location recognition and linking. Furthermore, a novel post-processing strategy, named Geographical Pair Linking, is developed to improve the linking performance. In this way, DLocRL does not require hand-crafted features. The experimental results show the effectiveness of DLocRL on fine-grained location recognition and linking with a real-world Twitter dataset. …

NAS-Bench-101
Recent advances in neural architecture search (NAS) demand tremendous computational resources. This makes it difficult to reproduce experiments and imposes a barrier-to-entry to researchers without access to large-scale computation. We aim to ameliorate these problems by introducing NAS-Bench-101, the first public architecture dataset for NAS research. To build NAS-Bench-101, we carefully constructed a compact, yet expressive, search space, exploiting graph isomorphisms to identify 423k unique convolutional architectures. We trained and evaluated all of these architectures multiple times on CIFAR-10 and compiled the results into a large dataset. All together, NAS-Bench-101 contains the metrics of over 5 million models, the largest dataset of its kind thus far. This allows researchers to evaluate the quality of a diverse range of models in milliseconds by querying the pre-computed dataset. We demonstrate its utility by analyzing the dataset as a whole and by benchmarking a range of architecture optimization algorithms. …

Constrained Recurrent Sparse Auto-Encoder (CRsAE)
Convolutional dictionary learning (CDL) has become a popular method for learning sparse representations from data. State-of-the-art algorithms perform dictionary learning (DL) through an optimization-based alternating-minimization procedure that comprises a sparse coding and a dictionary update step respectively. Here, we draw connections between CDL and neural networks by proposing an architecture for CDL termed the constrained recurrent sparse auto-encoder (CRsAE). We leverage the interpretation of the alternating-minimization algorithm for DL as an Expectation-Maximization algorithm to develop auto-encoders (AEs) that, for the first time, enable the simultaneous training of the dictionary and regularization parameter. The forward pass of the encoder, which performs sparse coding, solves the E-step using an encoding matrix and a soft-thresholding non-linearity imposed by the FISTA algorithm. The encoder in this regard is a variant of residual and recurrent neural networks. The M-step is implemented via a two-stage back-propagation. In the first stage, we perform back-propagation through the AE formed by the encoder and a linear decoder whose parameters are tied to the encoder. This stage parallels the dictionary update step in DL. In the second stage, we update the regularization parameter by performing back-propagation through the encoder using a loss function that includes a prior on the parameter motivated by Bayesian statistics. We leverage GPUs to achieve significant computational gains relative to state-of-the-art optimization-based approaches to CDL. We apply CRsAE to spike sorting, the problem of identifying the time of occurrence of neural action potentials in recordings of electrical activity from the brain. We demonstrate on recordings lasting hours that CRsAE speeds up spike sorting by 900x compared to notoriously slow classical algorithms based on convex optimization. …

Participatory Sensing
Participatory sensing is the concept of communities (or other groups of people) contributing sensory information to form a body of knowledge. …

Conceptual Interoperability Constraint (COIN)
Building meaningful interoperation with external software units requires performing the conceptual interoperability analysis that starts with identifying the conceptual interoperability constraints of each software unit, then it compares the systems’ constraints to detect their conceptual mismatch. We call the conceptual interoperability constraints (the COINs) that can be of different types including structure, dynamic, and quality. Missing such constraints may lead to unexpected mismatches, expensive resolution, and running-late projects. However, it is a challenging task for software architects and analysts to manually analyze the unstructured text in API documents to identify the COINs. Not only it is a tedious and time-consuming task, but also it needs knowledge about the constraint types. In this article, we present and evaluate our idea of utilizing machine learning techniques in automating the COIN identification, which is the first step of conceptual interoperability analysis, from human text in API documents. Our empirical research started with a multiple-case study to build the ground truth dataset, on which we contributed our machine learning COIN-Classification Model. We show the model’s robustness through experiments using different machine learning text-classification algorithms. The experiments’ results revealed that our model can achieve up to 87% accuracy in automatically identifying the COINs in text. Thus, we implemented a tool that embeds our model to demonstrate its practical value in industrial context. Then, we evaluated the practitioners’ acceptance for the tool and found that they significantly agreed on its usefulness and ease of use. …

Singular Vector Canonical Correlation Analysis (SVCCA)
We introduce a technique based on the singular vector canonical correlation analysis (SVCCA) for measuring the generality of neural network layers across a continuously-parametrized set of tasks. We illustrate this method by studying generality in neural networks trained to solve parametrized boundary value problems based on the Poisson partial differential equation. We find that the first hidden layer is general, and that deeper layers are successively more specific. Next, we validate our method against an existing technique that measures layer generality using transfer learning experiments. We find excellent agreement between the two methods, and note that our method is much faster, particularly for continuously-parametrized problems. Finally, we visualize the general representations of the first layers, and interpret them as generalized coordinates over the input domain. …

Collaborative GAN
Unlike a conventional background inpainting approach that infers a missing area from image patches similar to the background, face completion requires semantic knowledge about the target object for realistic outputs. Current image inpainting approaches utilize generative adversarial networks (GANs) to achieve such semantic understanding. However, in adversarial learning, the semantic knowledge is learned implicitly and hence good semantic understanding is not always guaranteed. In this work, we propose a collaborative adversarial learning approach to face completion to explicitly induce the training process. Our method is formulated under a novel generative framework called collaborative GAN (collaGAN), which allows better semantic understanding of a target object through collaborative learning of multiple tasks including face completion, landmark detection, and semantic segmentation. Together with the collaGAN, we also introduce an inpainting concentrated scheme such that the model emphasizes more on inpainting instead of autoencoding. Extensive experiments show that the proposed designs are indeed effective and collaborative adversarial learning provides better feature representations of the faces. In comparison with other generative image inpainting models and single task learning methods, our solution produces superior performances on all tasks. …

Collective And Point Anomalies (CAPA)
The challenge of efficiently identifying anomalies in data sequences is an important statistical problem that now arises in many applications. Whilst there has been substantial work aimed at making statistical analyses robust to outliers, or point anomalies, there has been much less work on detecting anomalous segments, or collective anomalies. By bringing together ideas from changepoint detection and robust statistics, we introduce Collective And Point Anomalies (CAPA), a computationally efficient approach that is suitable when collective anomalies are characterised by either a change in mean, variance, or both, and distinguishes them from point anomalies. Theoretical results establish the consistency of CAPA at detecting collective anomalies and empirical results show that CAPA has close to linear computational cost as well as being more accurate at detecting and locating collective anomalies than other approaches. We demonstrate the utility of CAPA through its ability to detect exoplanets from light curve data from the Kepler telescope. …

Algorithmic Complexity (AC)
The information content or complexity of an object can be measured by the length of its shortest description. For instance the string “01010101010101010101010101010101” has the short description “16 repetitions of 01”, while “11001000011000011101111011101100” presumably has no simpler description other than writing down the string itself. More formally, the Algorithmic “Kolmogorov” Complexity (AC) of a string x is defined as the length of the shortest program that computes or outputs x , where the program is run on some fixed reference universal computer. …

LISTwise ExplaiNer (LISTEN)
There is an increasing demand for algorithms to explain their outcomes. So far, there is no method that explains the rankings produced by a ranking algorithm. To address this gap we propose LISTEN, a LISTwise ExplaiNer, to explain rankings produced by a ranking algorithm. To efficiently use LISTEN in production, we train a neural network to learn the underlying explanation space created by LISTEN; we call this model Q-LISTEN. We show that LISTEN produces faithful explanations and that Q-LISTEN is able to learn these explanations. Moreover, we show that LISTEN is safe to use in a real world environment: users of a news recommendation system do not behave significantly differently when they are exposed to explanations generated by LISTEN instead of manually generated explanations. …

Bounded Dijkstra (BD)
The shortest path (SP) and shortest paths tree (SPT) problems arise both as direct applications and as subroutines of overlay algorithms solving more complex problems such as the constrained shortest path (CSP) or the constrained minimum Steiner tree (CMST) problems. Often, such algorithms do not use the result of an SP subroutine if its total cost is greater than a given bound. For example, for delay-constrained problems, paths resulting from a least-delay SP run and whose delay is greater than the delay constraint of the original problem are not used by the overlay algorithm to construct its solution. As a result of the existence of these bounds, and because the Dijkstra SP algorithm discovers paths in increasing order of cost, we can terminate the SP search earlier, i.e., once it is known that paths with a greater total cost will not be considered by the overlay algorithm. This early termination allows to reduce the runtime of the SP subroutine, thereby reducing the runtime of the overlay algorithm without impacting its final result. We refer to this adaptation of Dijkstra for centralized implementations as bounded Dijkstra (BD). On the example of CSP algorithms, we confirm the usefulness of BD by showing that it can reduce the runtime of some algorithms by 75% on average. …

Dynamic Vocabulary Sequence-to-Sequence (DVS2S)
We study response generation for open domain conversation in chatbots. Existing methods assume that words in responses are generated from an identical vocabulary regardless of their inputs, which not only makes them vulnerable to generic patterns and irrelevant noise, but also causes a high cost in decoding. We propose a dynamic vocabulary sequence-to-sequence (DVS2S) model which allows each input to possess their own vocabulary in decoding. In training, vocabulary construction and response generation are jointly learned by maximizing a lower bound of the true objective with a Monte Carlo sampling method. In inference, the model dynamically allocates a small vocabulary for an input with the word prediction model, and conducts decoding only with the small vocabulary. Because of the dynamic vocabulary mechanism, DVS2S eludes many generic patterns and irrelevant words in generation, and enjoys efficient decoding at the same time. Experimental results on both automatic metrics and human annotations show that DVS2S can significantly outperform state-of-the-art methods in terms of response quality, but only requires 60% decoding time compared to the most efficient baseline. …

Transfer Metric Learning (TML)
Distance metric learning (DML) aims to find an appropriate way to reveal the underlying data relationship. It is critical in many machine learning, pattern recognition and data mining algorithms, and usually require large amount of label information (class labels or pair/triplet constraints) to achieve satisfactory performance. However, the label information may be insufficient in real-world applications due to the high-labeling cost, and DML may fail in this case. Transfer metric learning (TML) is able to mitigate this issue for DML in the domain of interest (target domain) by leveraging knowledge/information from other related domains (source domains). Although achieved a certain level of development, TML has limited success in various aspects such as selective transfer, theoretical understanding, handling complex data, big data and extreme cases. In this survey, we present a systematic review of the TML literature. In particular, we group TML into different categories according to different settings and metric transfer strategies, such as direct metric approximation, subspace approximation, distance approximation, and distribution approximation. A summarization and insightful discussion of the various TML approaches and their applications will be presented. Finally, we provide some challenges and possible future directions. …

DNNVM
The convolutional neural network (CNN) has become a state-of-the-art method for several artificial intelligence domains in recent years. The increasingly complex CNN models are both computation-bound and I/O-bound. FPGA-based accelerators driven by custom instruction set architecture (ISA) achieve a balance between generality and efficiency, but there is much on them left to be optimized. We propose the full-stack compiler DNNVM, which is an integration of optimizers for graphs, loops and data layouts, and an assembler, a runtime supporter and a validation environment. The DNNVM works in the context of deep learning frameworks and transforms CNN models into the directed acyclic graph: XGraph. Based on XGraph, we transform the optimization challenges for both the data layout and pipeline into graph-level problems. DNNVM enumerates all potentially profitable fusion opportunities by a heuristic subgraph isomorphism algorithm to leverage pipeline and data layout optimizations, and searches for the optimal execution strategies of the whole computing graph. On the Xilinx ZU2 @330 MHz and ZU9 @330 MHz, we achieve equivalently state-of-the-art performance on our benchmarks by naive implementations without optimizations, and the throughput is further improved up to 1.26x by leveraging heterogeneous optimizations in DNNVM. Finally, with ZU9 @330 MHz, we achieve state-of-the-art performance for VGG and ResNet50. We achieve a throughput of 2.82 TOPs/s and an energy efficiency of 123.7 GOPs/s/W for VGG. Additionally, we achieve 1.38 TOPs/s for ResNet50. …

Deep Sequential Model for Knowledge Graph Completion (DSKG)
Knowledge graph (KG) completion aims to fill the missing facts in a KG, where a fact is represented as a triple in the form of $(subject, relation, object)$. Current KG completion models compel two-thirds of a triple provided (e.g., $subject$ and $relation$) to predict the remaining one. In this paper, we propose a new model, which uses a KG-specific multi-layer recurrent neutral network (RNN) to model triples in a KG as sequences. It outperformed several state-of-the-art KG completion models on the conventional entity prediction task for many evaluation metrics, based on two benchmark datasets and a more difficult dataset. Furthermore, our model is enabled by the sequential characteristic and thus capable of predicting the whole triples only given one entity. Our experiments demonstrated that our model achieved promising performance on this new triple prediction task. …

Coin.AI
One decade ago, Bitcoin was introduced, becoming the first cryptocurrency and establishing the concept of ‘blockchain’ as a distributed ledger. As of today, there are many different implementations of cryptocurrencies working over a blockchain, with different approaches and philosofies. However, many of them share one common feature: they require proof-of-work to support the generation of blocks (mining) and, eventually, the generation of money. This proof-of-work scheme often consists on the resolution of a cryptography problem, most commonly breaking a hash value, which can only be achieved through brute-force. The main drawback of proof-of-work is that it requires ridiculously large amounts of energy which do not have any useful outcome beyond supporting the currency. In this paper, we present a theoretical proposal that introduces a proof-of-useful-work scheme to support a cryptocurrency running over a blockchain, which we named Coin.AI. In this system, the mining scheme requires training deep learning models, and a block is only mined when the performance of such model exceeds a threshold. The distributed system allows for nodes to verify the models delivered by miners in an easy way (certainly much more efficiently than the mining process itself), determining when a block is to be generated. Additionally, this paper presents a proof-of-storage scheme for rewarding users that provide storage for the deep learning models, as well as a theoretical dissertation on how the mechanics of the system could be articulated with the ultimate goal of democratizing the access to artificial intelligence. …

Bottleneck Simulator
Deep reinforcement learning has recently shown many impressive successes. However, one major obstacle towards applying such methods to real-world problems is their lack of data-efficiency. To this end, we propose the Bottleneck Simulator: a model-based reinforcement learning method which combines a learned, factorized transition model of the environment with rollout simulations to learn an effective policy from few examples. The learned transition model employs an abstract, discrete (bottleneck) state, which increases sample efficiency by reducing the number of model parameters and by exploiting structural properties of the environment. We provide a mathematical analysis of the Bottleneck Simulator in terms of fixed points of the learned policy, which reveals how performance is affected by four distinct sources of error: an error related to the abstract space structure, an error related to the transition model estimation variance, an error related to the transition model estimation bias, and an error related to the transition model class bias. Finally, we evaluate the Bottleneck Simulator on two natural language processing tasks: a text adventure game and a real-world, complex dialogue response selection task. On both tasks, the Bottleneck Simulator yields excellent performance beating competing approaches. …

Condensed Representation
Correlated pattern mining has increasingly become an important task in data mining since these patterns allow conveying knowledge about meaningful and surprising relations among data. Frequent correlated patterns were thoroughly studied in the literature. In this thesis, we propose to benefit from both frequent correlated as well as rare correlated patterns according to the bond correlation measure. We propose to extract a subset without information loss of the sets of frequent correlated and of rare correlated patterns, this subset is called “Condensed Representation“. In this regard, we are based on the notions derived from the Formal Concept Analysis FCA, specifically the equivalence classes associated to a closure operator fbond dedicated to the bond measure, to introduce new concise representations of both frequent correlated and rare correlated patterns. …

Graph Feature Network (GFN)
Graph Neural Nets (GNNs) have received increasing attentions, partially due to their superior performance in many node and graph classification tasks. However, there is a lack of understanding on what they are learning and how sophisticated the learned graph functions are. In this work, we first propose Graph Feature Network (GFN), a simple lightweight neural net defined on a set of graph augmented features. We then propose a dissection of GNNs on graph classification into two parts: 1) the graph filtering, where graph-based neighbor aggregations are performed, and 2) the set function, where a set of hidden node features are composed for prediction. To test the importance of these two parts separately, we prove and leverage the connection that GFN can be derived by linearizing graph filtering part of GNN. Empirically we perform evaluations on common graph classification benchmarks. To our surprise, we find that, despite the simplification, GFN could match or exceed the best accuracies produced by recently proposed GNNs, with a fraction of computation cost. Our results provide new perspectives on both the functions that GNNs learned and the current benchmarks for evaluating them. …

Quantum Annealing (QA)
Quantum annealing (QA) is a metaheuristic for finding the global minimum of a given objective function over a given set of candidate solutions (candidate states), by a process using quantum fluctuations. Quantum annealing is used mainly for problems where the search space is discrete (combinatorial optimization problems) with many local minima; such as finding the ground state of a spin glass. It was formulated in its present form by T. Kadowaki and H. Nishimori (ja) in ‘Quantum annealing in the transverse Ising model’ though a proposal in a different form had been made by A. B. Finilla, M. A. Gomez, C. Sebenik and J. D. Doll, in ‘Quantum annealing: A new method for minimizing multidimensional functions’. Quantum annealing starts from a quantum-mechanical superposition of all possible states (candidate states) with equal weights. Then the system evolves following the time-dependent Schrödinger equation, a natural quantum-mechanical evolution of physical systems. The amplitudes of all candidate states keep changing, realizing a quantum parallelism, according to the time-dependent strength of the transverse field, which causes quantum tunneling between states. If the rate of change of the transverse field is slow enough, the system stays close to the ground state of the instantaneous Hamiltonian, i.e., adiabatic quantum computation. If the rate of change of the transverse field is accelerated, the system may leave the ground state temporarily but produce a higher likelihood of concluding in the ground state of the final problem Hamiltonian, i.e., diabatic quantum computation. The transverse field is finally switched off, and the system is expected to have reached the ground state of the classical Ising model that corresponds to the solution to the original optimization problem. An experimental demonstration of the success of quantum annealing for random magnets was reported immediately after the initial theoretical proposal. An introduction to combinatorial optimization (NP-hard) problems, the general structure of quantum annealing-based algorithms and two examples of this kind of algorithms for solving instances of the max-SAT and Minimum Multicut problems together with an overview of the quantum annealing systems manufactured by D-Wave Systems are presented in. …

Robust Optimization
Robust optimization is a field of optimization theory that deals with optimization problems in which a certain measure of robustness is sought against uncertainty that can be represented as deterministic variability in the value of the parameters of the problem itself and/or its solution. …

Deadline-Aware Task rEplication for Vehicular Cloud (DATE-V)
Vehicular Cloud Computing (VCC) is a new technological shift which exploits the computation and storage resources on vehicles for computational service provisioning. Spare on-board resources are pooled by a VCC operator, e.g. a roadside unit, to complete task requests using the vehicle-as-a-resource framework. In this paper, we investigate timely service provisioning for deadline-constrained tasks in VCC systems by leveraging the task replication technique (i.e., allowing one task to be executed by several server vehicles). A learning-based algorithm, called DATE-V (Deadline-Aware Task rEplication for Vehicular Cloud), is proposed to address the special issues in VCC systems including uncertainty of vehicle movements, volatile vehicle members, and large vehicle population. The proposed algorithm is developed based on a novel Contextual-Combinatorial Multi-Armed Bandit (CC-MAB) learning framework. DATE-V is `contextual’ because it utilizes side information (context) of vehicles and tasks to infer the completion probability of a task replication under random vehicle movements. DATE-V is `combinatorial’ because it aims to replicate the received task and send the task replications to multiple server vehicles to guarantee the service timeliness. We rigorously prove that our learning algorithm achieves a sublinear regret bound compared to an oracle algorithm that knows the exact completion probability of any task replications. Simulations are carried out based on real-world vehicle movement traces and the results show that DATE-V significantly outperforms benchmark solutions. …

Language-Conditioned Graph Network (LCGN)
Solving grounded language tasks often requires reasoning about relationships between objects in the context of a given task. For example, to answer the question “What color is the mug on the plate?” we must check the color of the specific mug that satisfies the “on” relationship with respect to the plate. Recent work has proposed various methods capable of complex relational reasoning. However, most of their power is in the inference structure, while the scene is represented with simple local appearance features. In this paper, we take an alternate approach and build contextualized representations for objects in a visual scene to support relational reasoning. We propose a general framework of Language-Conditioned Graph Networks (LCGN), where each node represents an object, and is described by a context-aware representation from related objects through iterative message passing conditioned on the textual input. E.g., conditioning on the “on” relationship to the plate, the object “mug” gathers messages from the object “plate” to update its representation to “mug on the plate”, which can be easily consumed by a simple classifier for answer prediction. We experimentally show that our LCGN approach effectively supports relational reasoning and improves performance across several tasks and datasets. …

DeepFirearm
There are great demands for automatically regulating inappropriate appearance of shocking firearm images in social media or identifying firearm types in forensics. Image retrieval techniques have great potential to solve these problems. To facilitate research in this area, we introduce Firearm 14k, a large dataset consisting of over 14,000 images in 167 categories. It can be used for both fine-grained recognition and retrieval of firearm images. Recent advances in image retrieval are mainly driven by fine-tuning state-of-the-art convolutional neural networks for retrieval task. The conventional single margin contrastive loss, known for its simplicity and good performance, has been widely used. We find that it performs poorly on the Firearm 14k dataset due to: (1) Loss contributed by positive and negative image pairs is unbalanced during training process. (2) A huge domain gap exists between this dataset and ImageNet. We propose to deal with the unbalanced loss by employing a double margin contrastive loss. We tackle the domain gap issue with a two-stage training strategy, where we first fine-tune the network for classification, and then fine-tune it for retrieval. Experimental results show that our approach outperforms the conventional single margin approach by a large margin (up to 88.5% relative improvement) and even surpasses the strong triplet-loss-based approach. …

Easy Positive Triplet Mining
Deep metric learning seeks to define an embedding where semantically similar images are embedded to nearby locations, and semantically dissimilar images are embedded to distant locations. Substantial work has focused on loss functions and strategies to learn these embeddings by pushing images from the same class as close together in the embedding space as possible. In this paper, we propose an alternative, loosened embedding strategy that requires the embedding function only map each training image to the most similar examples from the same class, an approach we call ‘Easy Positive’ mining. We provide a collection of experiments and visualizations that highlight that this Easy Positive mining leads to embeddings that are more flexible and generalize better to new unseen data. This simple mining strategy yields recall performance that exceeds state of the art approaches (including those with complicated loss functions and ensemble methods) on image retrieval datasets including CUB, Stanford Online Products, In-Shop Clothes and Hotels-50K. …

Generalized Dual Averaging
We present a new class of algorithms for solving regularized optimization and saddle point problems. We analyse this class of methods for convex optimization and convex-concave saddle point problems and expect that they will be useful for solving non-convex problems as well. For convex and convex-concave problems, our algorithms form a novel class of primal dual subgradient methods. This new class of methods extends existing methods by utilizing a more general bound on the objective error and duality gap. This leads to methods for which we can control the step size of the proximal update, which is of interest for problems where the sparsity of the iterates is important. We prove that our class of methods is optimal from the point of view of worst-case black-box complexity for convex optimization problems, and derive a version for convex-concave saddle point problems. We also analyse our methods in the stochastic and online settings. Finally, we exhibit a variety of special cases and discuss their usefulness for non-convex optimization. …

SPFlow
We introduce SPFlow, an open-source Python library providing a simple interface to inference, learning and manipulation routines for deep and tractable probabilistic models called Sum-Product Networks (SPNs). The library allows one to quickly create SPNs both from data and through a domain specific language (DSL). It efficiently implements several probabilistic inference routines like computing marginals, conditionals and (approximate) most probable explanations (MPEs) along with sampling as well as utilities for serializing, plotting and structure statistics on an SPN. Moreover, many of the algorithms proposed in the literature to learn the structure and parameters of SPNs are readily available in SPFlow. Furthermore, SPFlow is extremely extensible and customizable, allowing users to promptly distill new inference and learning routines by injecting custom code into a lightweight functional-oriented API framework. This is achieved in SPFlow by keeping an internal Python representation of the graph structure that also enables practical compilation of an SPN into a TensorFlow graph, C, CUDA or FPGA custom code, significantly speeding-up computations. …

Generative Adversarial Imputation Net (GAIN)
We propose a novel method for imputing missing data by adapting the well-known Generative Adversarial Nets (GAN) framework. Accordingly, we call our method Generative Adversarial Imputation Nets (GAIN). The generator (G) observes some components of a real data vector, imputes the missing components conditioned on what is actually observed, and outputs a completed vector. The discriminator (D) then takes a completed vector and attempts to determine which components were actually observed and which were imputed. To ensure that D forces G to learn the desired distribution, we provide D with some additional information in the form of a hint vector. The hint reveals to D partial information about the missingness of the original sample, which is used by D to focus its attention on the imputation quality of particular components. This hint ensures that G does in fact learn to generate according to the true data distribution. We tested our method on various datasets and found that GAIN significantly outperforms state-of-the-art imputation methods. …

Multivariate Bernoulli Autoregressive Process (BAR)
Multivariate Bernoulli autoregressive (BAR) processes model time series of events in which the likelihood of current events is determined by the times and locations of past events. These processes can be used to model nonlinear dynamical systems corresponding to criminal activity, responses of patients to different medical treatment plans, opinion dynamics across social networks, epidemic spread, and more. Past work examines this problem under the assumption that the event data is complete, but in many cases only a fraction of events are observed. Incomplete observations pose a significant challenge in this setting because the unobserved events still govern the underlying dynamical system. In this work, we develop a novel approach to estimating the parameters of a BAR process in the presence of unobserved events via an unbiased estimator of the complete data log-likelihood function. We propose a computationally efficient estimation algorithm which approximates this estimator via Taylor series truncation and establish theoretical results for both the statistical error and optimization error of our algorithm. We further justify our approach by testing our method on both simulated data and a real data set consisting of crimes recorded by the city of Chicago. …

Attention-Aware Generative Adversarial Network (ATA-GAN)
In this work, we present a novel approach for training Generative Adversarial Networks (GANs). Using the attention maps produced by a Teacher- Network we are able to improve the quality of the generated images as well as perform weakly object localization on the generated images. To this end, we generate images of HEp-2 cells captured with Indirect Imunofluoresence (IIF) and study the ability of our network to perform a weakly localization of the cell. Firstly, we demonstrate that whilst GANs can learn the mapping between the input domain and the target distribution efficiently, the discriminator network is not able to detect the regions of interest. Secondly, we present a novel attention transfer mechanism which allows us to enforce the discriminator to put emphasis on the regions of interest via transfer learning. Thirdly, we show that this leads to more realistic images, as the discriminator learns to put emphasis on the area of interest. Fourthly, the proposed method allows one to generate both images as well as attention maps which can be useful for data annotation e.g in object detection. …

Attentive Adversarial Domain-Invariant Training (AADIT)
Adversarial domain-invariant training (ADIT) proves to be effective in suppressing the effects of domain variability in acoustic modeling and has led to improved performance in automatic speech recognition (ASR). In ADIT, an auxiliary domain classifier takes in equally-weighted deep features from a deep neural network (DNN) acoustic model and is trained to improve their domain-invariance by optimizing an adversarial loss function. In this work, we propose an attentive Adversarial domain-invariant training (AADIT) in which we advance the domain classifier with an attention mechanism to automatically weight the input deep features according to their importance in domain classification. With this attentive re-weighting, AADIT can focus on the domain normalization of phonetic components that are more susceptible to domain variability and generates deep features with improved domain-invariance and senone-discriminativity over ADIT. Most importantly, the attention block serves only as an external component to the DNN acoustic model and is not involved in ASR, so AADIT can be used to improve the acoustic modeling with any DNN architectures. More generally, the same methodology can improve any adversarial learning system with an auxiliary discriminator. Evaluated on CHiME-3 dataset, the AADIT achieves 13.6% and 9.3% relative WER improvements, respectively, over a multi-conditional model and a strong ADIT baseline. …

Markov Modulated Hawkes Process (MMHP)
Modeling event dynamics is central to many disciplines. Patterns in observed event arrival times are commonly modeled using point processes. Such event arrival data often exhibits self-exciting, heterogeneous and sporadic trends, which is challenging for conventional models. It is reasonable to assume that there exists a hidden state process that drives different event dynamics at different states. In this paper, we propose a Markov Modulated Hawkes Process (MMHP) model for learning such a mixture of event dynamics and develop corresponding inference algorithms. Numerical experiments using synthetic data and data from an animal behavior study demonstrate that MMHP with the proposed estimation algorithms consistently recover the true hidden state process in simulations, and separately captures distinct event dynamics that reveal interesting social structures in the real data. …