Multimodal language analysis is an emerging research area in natural language processing that models language in a multimodal manner. It aims to understand language from the modalities of text, visual, and acoustic by modeling both intra-modal and cross-modal interactions. BERT (Bidirectional Encoder Representations from Transformers) provides strong contextual language representations after training on large-scale unlabeled corpora. Fine-tuning the vanilla BERT model has shown promising results in building state-of-the-art models for diverse NLP tasks like question answering and language inference. However, fine-tuning BERT in the presence of information from other modalities remains an open research problem. In this paper, we inject multimodal information within the input space of BERT network for modeling multimodal language. The proposed injection method allows BERT to reach a new state of the art of $84.38\%$ binary accuracy on CMU-MOSI dataset (multimodal sentiment analysis) with a gap of 5.98 percent to the previous state of the art and 1.02 percent to the text-only BERT.

In this paper, we propose a new method to build fair Neural-Network classifiers by using a constraint based on the Wasserstein distance. More specifically, we detail how to efficiently compute the gradients of Wasserstein-2 regularizers for Neural-Networks. The proposed strategy is then used to train Neural-Networks decision rules which favor fair predictions. Our method fully takes into account two specificities of Neural-Networks training: (1) The network parameters are indirectly learned based on automatic differentiation and on the loss gradients, and (2) batch training is the gold standard to approximate the parameter gradients, as it requires a reasonable amount of computations and it can efficiently explore the parameters space. Results are shown on synthetic data, as well as on the UCI Adult Income Dataset. Our method is shown to perform well compared with 'ZafarICWWW17' and linear-regression with Wasserstein-1 regularization, as in 'JiangUAI19', in particular when non-linear decision rules are required for accurate predictions.

PREPRINT 1 MimickNet, Matching Clinical Post-Processing Under Realistic Black-Box Constraints Ouwen Huang, Student Member, IEEE, Will Long, Student Member, IEEE, Nick Bottenus, Gregg E. Trahey, Senior Member, IEEE, Sina Farsiu, Senior Member, IEEE, and Mark L. Palmeri, Senior Member, IEEE Abstract --Image post-processing is used in clinical-grade ultrasound scanners to improve image quality (e.g., reduce speckle noise and enhance contrast). These post-processing techniques vary across manufacturers and are generally kept proprietary, which presents a challenge for researchers looking to match current clinical-grade workflows. We introduce a deep learning framework, MimickNet, that transforms raw conventional delay-and-summed (DAS) beams into the approximate post-processed images found on clinical-grade scanners. Training MimickNet only requires post-processed image samples from a scanner of interest without the need for explicit pairing to raw DAS data. This flexibility allows it to hypothetically approximate any manufacturer's post-processing without access to the pre-processed data. MimickNet generates images with an average similarity index measurement (SSIM) of 0.930 0.0892 on a 300 cineloop test set, and it generalizes to cardiac cineloops outside of our train-test distribution achieving an SSIM of 0.967 0.002. We also explore the theoretical SSIM achievable by evaluating MimickNet performance when trained under gray-box constraints (i.e., when both pre-processed and post-processed images are available). T o our knowledge, this is the first work to establish deep learning models that closely approximate current clinical-grade ultrasound post-processing under realistic black-box constraints where before and after post-processing data is unavailable. MimickNet serves as a clinical post-processing baseline for future works in ultrasound image formation to compare against.

Min-max optimization has attracted much attention in the machine learning community due to the popularization of deep generative models and adversarial training. The optimization is quite different from traditional minimization analysis. For example, gradient descent does not converge in one of the simplest settings -- bilinear games. In this paper, we try to understand several gradient-based algorithms for bilinear min-max games: gradient descent, extra-gradient, optimistic gradient descent and the momentum method, for both simultaneous and alternating updates. We provide necessary and sufficient conditions for their convergence, with the Schur theorem. Furthermore, by extending these algorithms to more general parameter settings, we are able to optimize over larger parameter spaces to find the optimal convergence rates. Our results imply that alternating updates converge more easily in min-max games than simultaneous updates.

Department of Pharmaceutical Biosciences Uppsala University, Uppsala, Sweden Abstract Conformal Prediction is a framework that produces prediction intervals based on the output from a machine learning algorithm. In this paper we explore the case when training data is made up of multiple parts available in different sources that cannot be pooled. We here consider the regression case and propose a method where a conformal predictor is trained on each data source independently, and where the prediction intervals are then combined into a single interval. We call the approach Non-Disclosed Conformal Prediction (NDCP), and we evaluate it on a regression dataset from the UCI machine learning repository using support vector regression as the underlying machine learning algorithm, with varying number of data sources and sizes. The results show that the proposed method produces conservatively valid prediction intervals, and while we cannot retain the same efficiency as when all data is used, efficiency is improved through the proposed approach as compared to predicting using a single arbitrarily chosen source.

--There has been considerable growth and interest in industrial applications of machine learning (ML) in recent years. ML engineers, as a consequence, are in high demand across the industry, yet improving the efficiency of ML engineers remains a fundamental challenge. Automated machine learning (AutoML) has emerged as a way to save time and effort on repetitive tasks in ML pipelines, such as data pre-processing, feature engineering, model selection, hyperparameter optimization, and prediction result analysis. In this paper, we investigate the current state of AutoML tools aiming to automate these tasks. We conduct various evaluations of the tools on many datasets, in different data segments, to examine their performance, and compare their advantages and disadvantages on different test cases. Automated Machine Learning (AutoML) promises major productivity boosts for data scientists, ML engineers and ML researchers by reducing repetitive tasks in machine learning pipelines. There are currently a number of different tools and platforms (both open-source and commercially available solutions) that try to automate these tasks. The goal of this paper is to address the following questions: (i) what are the available ML functionalities provided by the tools; (ii) how the tools perform when facing a wide spectrum of real world datasets; (iii) find the tradeoff between optimization speed and accuracy of the results; and (iv) the reproducibility of the results (a.k.a.

Automatic segmentation methods based on deep learning have recently demonstrated state-of-the-art performance, outperforming the ordinary methods. Nevertheless, these methods are inapplicable for small datasets, which are very common in medical problems. To this end, we propose a knowledge transfer method between diseases via the Generative Bayesian Prior network. Our approach is compared to a pre-train approach and random initialization and obtains the best results in terms of Dice Similarity Coefficient metric for the small subsets of the Brain Tumor Segmentation 2018 database (BRATS2018).

The collaborative ranking problem has been an important open research question as most recommendation problems can be naturally formulated as ranking problems. While much of collaborative ranking methodology assumes static ranking data, the importance of temporal information to improving ranking performance is increasingly apparent. Recent advances in deep learning, especially the discovery of various attention mechanisms and newer architectures in addition to widely used RNN and CNN in natural language processing, have allowed us to make better use of the temporal ordering of items that each user has engaged with. In particular, the SASRec model, inspired by the popular Transformer model in natural languages processing, has achieved state-of-art results in the temporal collaborative ranking problem and enjoyed more than 10x speed-up when compared to earlier CNN/RNN-based methods. However, SASRec is inherently an un-personalized model and does not include personalized user embeddings. To overcome this limitation, we propose a Personalized Transformer (SSE-PT) model, outperforming SASRec by almost 5% in terms of NDCG@10 on 5 real-world datasets. Furthermore, after examining some random users' engagement history and corresponding attention heat maps used during the inference stage, we find our model is not only more interpretable but also able to focus on recent engagement patterns for each user. Moreover, our SSE-PT model with a slight modification, which we call SSE-PT++, can handle extremely long sequences and outperform SASRec in ranking results with comparable training speed, striking a balance between performance and speed requirements. Code and data are open sourced at https://github.com/wuliwei9278/SSE-PT.

Sex trafficking is a global epidemic. Escort websites are a primary vehicle for selling the services of such trafficking victims and thus a major driver of trafficker revenue. Many law enforcement agencies do not have the resources to manually identify leads from the millions of escort ads posted across dozens of public websites. We propose an ordinal regression neural network to identify escort ads that are likely linked to sex trafficking. Our model uses a modified cost function to mitigate inconsistencies in predictions often associated with nonparametric ordinal regression and leverages recent advancements in deep learning to improve prediction accuracy. The proposed method significantly improves on the previous state-of-the-art on Trafficking-10K, an expert-annotated dataset of escort ads. Additionally, because traffickers use acronyms, deliberate typographical errors, and emojis to replace explicit keywords, we demonstrate how to expand the lexicon of trafficking flags through word embeddings and t-SNE.

Network alignment is a critical task to a wide variety of fields. Many existing works leverage on representation learning to accomplish this task without eliminating domain representation bias induced by domain-dependent features, which yield inferior alignment performance. This paper proposes a unified deep architecture (DANA) to obtain a domain-invariant representation for network alignment via an adversarial domain classifier. Specifically, we employ the graph convolutional networks to perform network embedding under the domain adversarial principle, given a small set of observed anchors. Then, the semi-supervised learning framework is optimized by maximizing a posterior probability distribution of observed anchors and the loss of a domain classifier simultaneously. We also develop a few variants of our model, such as, direction-aware network alignment, weight-sharing for directed networks and simplification of parameter space. Experiments on three real-world social network datasets demonstrate that our proposed approaches achieve state-of-the-art alignment results.