A BSTRACT V ariance reduction methods such as SVRG (Johnson & Zhang, 2013) and SpiderBoost (Wang et al., 2018) use a mixture of large and small batch gradients to reduce the variance of stochastic gradients. Compared to SGD (Robbins & Monro, 1951), these methods require at least double the number of operations per update to model parameters. To reduce the computational cost of these methods, we introduce a new sparsity operator: The random-top- k operator. Our operator reduces computational complexity by estimating gradient sparsity exhibited in a variety of applications by combining the top-k operator (Stich et al., 2018; Aji & Heafield, 2017) and the randomized coordinate descent operator. With this operator, large batch gradients offer an extra benefit beyond variance reduction: A reliable estimate of gradient sparsity. Theoretically, our algorithm is at least as good as the best algorithm (SpiderBoost), and further excels in performance whenever the random-top- k operator captures gradient sparsity. Empirically, our algorithm consistently outperforms SpiderBoost using various models on various tasks including image classification, natural language processing, and sparse matrix factorization. We also provide empirical evidence to support the intuition behind our algorithm via a simple gradient entropy computation, which serves to quantify gradient sparsity at every iteration. It updates the iterate x with x η f I(x), where η is the learning rate and f I(x) is the batch stochastic gradient, i.e. f I(x) 1 I null i I f i(x).

This work presents a two-stage neural architecture for learning and refining structural correspondences between graphs. First, we use localized node embeddings computed by a graph neural network to obtain an initial ranking of soft correspondences between nodes. Secondly, we employ synchronous message passing networks to iteratively re-rank the soft correspondences to reach a matching consensus in local neighborhoods between graphs. We show, theoretically and empirically, that our message passing scheme computes a well-founded measure of consensus for corresponding neighborhoods, which is then used to guide the iterative re-ranking process. Our purely local and sparsity-aware architecture scales well to large, real-world inputs while still being able to recover global correspondences consistently. We demonstrate the practical effectiveness of our method on real-world tasks from the fields of computer vision and entity alignment between knowledge graphs, on which we improve upon the current state-of-the-art. Our source code is available under https://github.com/rusty1s/ deep-graph-matching-consensus.

In surface mount technology (SMT), mounted components on soldered pads are subject to move during reflow process. This capability is known as self-alignment and is the result of fluid dynamic behaviour of molten solder paste. This capability is critical in SMT because inaccurate self-alignment causes defects such as overhanging, tombstoning, etc. while on the other side, it can enable components to be perfectly self-assembled on or near the desire position. The aim of this study is to develop a machine learning model that predicts the components movement during reflow in x and y-directions as well as rotation. Our study is composed of two steps: (1) experimental data are studied to reveal the relationships between self-alignment and various factors including component geometry, pad geometry, etc. (2) advanced machine learning prediction models are applied to predict the distance and the direction of components shift using support vector regression (SVR), neural network (NN), and random forest regression (RFR). As a result, RFR can predict components shift with the average fitness of 99%, 99%, and 96% and with average prediction error of 13.47 (um), 12.02 (um), and 1.52 (deg.) for component shift in x, y, and rotational directions, respectively. This enhancement provides the future capability of the parameters' optimization in the pick and placement machine to control the best placement location and minimize the intrinsic defects caused by the self-alignment.

Surface mount technology (SMT) is an enhanced method in electronic packaging in which electronic components are placed directly on soldered printing circuit board (PCB) and are permanently attached on PCB with the aim of reflow soldering process. During reflow process, once deposited solder pastes start melting, electronic components move in a direction that achieve their highest symmetry. This motion is known as self-alignment since can correct potential mounting misalignment. In this study, two noticeable machine learning algorithms, including support vector regression (SVR) and random forest regression (RFR) are proposed as a prediction technique to (1) diagnose the relation among component self-alignment, deposited solder paste status and placement machining parameters, (2) predict the final component position on PCB in x, y, and rotational directions before entering in the reflow process. Based on the prediction result, a non-linear optimization model (NLP) is developed to optimize placement parameters at initial stage. Resultantly, RFR outperforms in terms of prediction model fitness and error. The optimization model is run for 6 samples in which the minimum Euclidean distance from component position after reflow process from ideal position (i.e., the center of pads) is outlined as 25.57 ({\mu}m) regarding defined boundaries in model.

With the explosive growth of online products and content, recommendation techniques have been considered as an effective tool to overcome information overload, improve user experience, and boost business revenue. In recent years, we have observed a new desideratum of considering long-term rewards of multiple related recommendation tasks simultaneously. The consideration of long-term rewards is strongly tied to business revenue and growth. Learning multiple tasks simultaneously could generally improve the performance of individual task due to knowledge sharing in multi-task learning. While a few existing works have studied long-term rewards in recommendations, they mainly focus on a single recommendation task. In this paper, we propose {\it PoDiRe}: a \underline{po}licy \underline{di}stilled \underline{re}commender that can address long-term rewards of recommendations and simultaneously handle multiple recommendation tasks. This novel recommendation solution is based on a marriage of deep reinforcement learning and knowledge distillation techniques, which is able to establish knowledge sharing among different tasks and reduce the size of a learning model. The resulting model is expected to attain better performance and lower response latency for real-time recommendation services. In collaboration with Samsung Game Launcher, one of the world's largest commercial mobile game platforms, we conduct a comprehensive experimental study on large-scale real data with hundreds of millions of events and show that our solution outperforms many state-of-the-art methods in terms of several standard evaluation metrics.

Anomaly detection algorithms find extensive use in various fields. This area of research has recently made great advances thanks to deep learning. A recent method, the deep Support Vector Data Description (deep SVDD), which is inspired by the classic kernel-based Support Vector Data Description (SVDD), is capable of simultaneously learning a feature representation of the data and a data-enclosing hypersphere. The method has shown promising results in both unsupervised and semi-supervised settings. However, deep SVDD suffers from hypersphere collapse---also known as mode collapse---, if the architecture of the model does not comply with certain architectural constraints, e.g. the removal of bias terms. These constraints limit the adaptability of the model and in some cases, may affect the model performance due to learning sub-optimal features. In this work, we consider two regularizers to prevent hypersphere collapse in deep SVDD. The first regularizer is based on injecting random noise via the standard cross-entropy loss. The second regularizer penalizes the minibatch variance when it becomes too small. Moreover, we introduce an adaptive weighting scheme to control the amount of penalization between the SVDD loss and the respective regularizer. Our proposed regularized variants of deep SVDD show encouraging results and outperform a prominent state-of-the-art method on a setup where the anomalies have no apparent geometrical structure.

Peer Assessment is a task of analysis and commenting on student's writing by peers, is core of all educational components both in campus and in MOOC's. However, with the sheer scale of MOOC's & its inherent personalised open ended learning, automatic grading and tools assisting grading at scale is highly important. Previously we presented survey on tasks of post classification, knowledge tracing and ended with brief review on Peer Assessment (PA), with some initial problems. In this review we shall continue review on PA from perspective of improving the review process itself. As such rest of this review focus on three facets of PA namely Auto grading and Peer Assessment Tools (we shall look only on how peer reviews/auto-grading is carried), strategies to handle Rogue Reviews, Peer Review Improvement using Natural Language Processing. The consolidated set of papers and resources so used are released in https://github.com/manikandan-ravikiran/cs6460-Survey-2.

In cyberattack detection and prevention systems, cybersecurity analysts always prefer solutions that are as interpretable and understandable as rule-based or signature-based detection. This is because of the need to tune and optimize these solutions to mitigate and control the effect of false positives and false negatives. Interpreting machine learning models is a new and open challenge. However, it is expected that an interpretable machine learning solution will be domain-specific. For instance, interpretable solutions for machine learning models in healthcare are different than solutions in malware detection. This is because the models are complex, and most of them work as a black-box. Recently, the increased ability for malware authors to bypass antimalware systems has forced security specialists to look to machine learning for creating robust detection systems. If these systems are to be relied on in the industry, then, among other challenges, they must also explain their predictions. The objective of this paper is to evaluate the current state-of-the-art ML models interpretability techniques when applied to ML-based malware detectors. We demonstrate interpretability techniques in practice and evaluate the effectiveness of existing interpretability techniques in the malware analysis domain.

Conversation s contain a wide spectrum of multimodal information that gives us hints about the emotions and moods of the speaker. In this paper, we developed a system that supports humans to analyze conversations. O ur main contribution is the identification of appropriat e multimodal features and the integration of such features into verbatim conversation transcripts . We demonstrate the ability of our system to take in a wide range of multimodal information and automatically generated a prediction score for the depression state of the individual. Our experiments showed that this approach yielded better performance than the baseline model . Furthermore, the multimodal narrative approach makes it easy to integrate learnings from other disciplines, such as conversational analys is and psychology. Lastly, this interdisciplinary and automated approach is a step towards emulating how practitioners record the course of treatment as well as emulating how conversational analysts have been analyzing conversations by hand.

This is where one needs to be careful. Our instinct might be to simply exponentiate the log-scale predictions back to raw-scale y. But our instinct would be wrong. Let's see why that is. If you like, you can skip the little bit of math that follows and scroll down to the section called Duan's smearing estimator.