Cold-start and sparsity problem are two key intrinsic problems to recommender systems. During the past two decades, researchers and industrial practitioners have spent considerable amount of efforts trying to solve the problems. However, for cold-start problem, most research relies on importing side information to transfer knowledge. A notable exception is ZeroMat, which uses no extra input data. Sparsity is a lesser noticed problem. In this paper, we propose a new algorithm named DotMat that relies on no extra input data, but is capable of solving cold-start and sparsity problems. In experiments, we prove that like ZeroMat, DotMat can achieve competitive results with recommender systems with full data, such as the classic matrix factorization algorithm.
Digital holography is a 3D imaging technique by emitting a laser beam with a plane wavefront to an object and measuring the intensity of the diffracted waveform, called holograms. The object's 3D shape can be obtained by numerical analysis of the captured holograms and recovering the incurred phase. Recently, deep learning (DL) methods have been used for more accurate holographic processing. However, most supervised methods require large datasets to train the model, which is rarely available in most DH applications due to the scarcity of samples or privacy concerns. A few one-shot DL-based recovery methods exist with no reliance on large datasets of paired images. Still, most of these methods often neglect the underlying physics law that governs wave propagation. These methods offer a black-box operation, which is not explainable, generalizable, and transferrable to other samples and applications. In this work, we propose a new DL architecture based on generative adversarial networks that uses a discriminative network for realizing a semantic measure for reconstruction quality while using a generative network as a function approximator to model the inverse of hologram formation. We impose smoothness on the background part of the recovered image using a progressive masking module powered by simulated annealing to enhance the reconstruction quality. The proposed method is one of its kind that exhibits high transferability to similar samples, which facilitates its fast deployment in time-sensitive applications without the need for retraining the network. The results show a considerable improvement to competitor methods in reconstruction quality (about 5 dB PSNR gain) and robustness to noise (about 50% reduction in PSNR vs noise increase rate).
Regularization is a popular technique to solve the overfitting problem of machine learning algorithms. Most regularization technique relies on parameter selection of the regularization coefficient. Plug-in method and cross-validation approach are two most common parameter selection approaches for regression methods such as Ridge Regression, Lasso Regression and Kernel Regression. Matrix factorization based recommendation system also has heavy reliance on the regularization technique. Most people select a single scalar value to regularize the user feature vector and item feature vector independently or collectively. In this paper, we prove that such approach of selecting regularization coefficient is invalid, and we provide a theoretically accurate method that outperforms the most widely used approach in both accuracy and fairness metrics.
This work provides the exact expression of the probability distribution of the hypervolume improvement (HVI) for bi-objective generalization of Bayesian optimization. Here, instead of a single-objective improvement, we consider the improvement of the hypervolume indicator concerning the current best approximation of the Pareto front. Gaussian process regression models are trained independently on both objective functions, resulting in a bi-variate separated Gaussian distribution serving as a predictive model for the vector-valued objective function. Some commonly HVI-based acquisition functions (probability of improvement and upper confidence bound) are also leveraged with the help of the exact distribution of HVI. In addition, we show the superior numerical accuracy and efficiency of the exact distribution compared to the commonly used approximation by Monte-Carlo sampling. Finally, we benchmark distribution-leveraged acquisition functions on the widely applied ZDT problem set, demonstrating a significant advantage of using the exact distribution of HVI in multi-objective Bayesian optimization.
With the rise of telemedicine, the task of developing Dialogue Systems for Medical Diagnosis (DSMD) has received much attention in recent years. Different from early researches that needed to rely on extra human resources and expertise to help construct the system, recent researches focused on how to build DSMD in a purely data-driven manner. However, the previous data-driven DSMD methods largely overlooked the system interpretability, which is critical for a medical application, and they also suffered from the data sparsity issue at the same time. In this paper, we explore how to bring interpretability to data-driven DSMD. Specifically, we propose a more interpretable decision process to implement the dialogue manager of DSMD by reasonably mimicking real doctors' inquiry logics, and we devise a model with highly transparent components to conduct the inference. Moreover, we collect a new DSMD dataset, which has a much larger scale, more diverse patterns and is of higher quality than the existing ones. The experiments show that our method obtains 7.7%, 10.0%, 3.0% absolute improvement in diagnosis accuracy respectively on three datasets, demonstrating the effectiveness of its rational decision process and model design. Our codes and the GMD-12 dataset are available at https://github.com/lwgkzl/BR-Agent.
Bayesian Optimization (BO) is a surrogate-based global optimization strategy that relies on a Gaussian Process regression (GPR) model to approximate the objective function and an acquisition function to suggest candidate points. It is well-known that BO does not scale well for high-dimensional problems because the GPR model requires substantially more data points to achieve sufficient accuracy and acquisition optimization becomes computationally expensive in high dimensions. Several recent works aim at addressing these issues, e.g., methods that implement online variable selection or conduct the search on a lower-dimensional sub-manifold of the original search space. Advancing our previous work of PCA-BO that learns a linear sub-manifold, this paper proposes a novel kernel PCA-assisted BO (KPCA-BO) algorithm, which embeds a non-linear sub-manifold in the search space and performs BO on this sub-manifold. Intuitively, constructing the GPR model on a lower-dimensional sub-manifold helps improve the modeling accuracy without requiring much more data from the objective function. Also, our approach defines the acquisition function on the lower-dimensional sub-manifold, making the acquisition optimization more manageable. We compare the performance of KPCA-BO to the vanilla BO and PCA-BO on the multi-modal problems of the COCO/BBOB benchmark suite. Empirical results show that KPCA-BO outperforms BO in terms of convergence speed on most test problems, and this benefit becomes more significant when the dimensionality increases. For the 60D functions, KPCA-BO surpasses PCA-BO in many test cases. Moreover, it efficiently reduces the CPU time required to train the GPR model and optimize the acquisition function compared to the vanilla BO.
Matrix Factorization is a widely adopted technique in the field of recommender system. Matrix Factorization techniques range from SVD, LDA, pLSA, SVD++, MatRec, Zipf Matrix Factorization and Item2Vec. In recent years, distributed word embeddings have inspired innovation in the area of recommender systems. Word2vec and GloVe have been especially emphasized in many industrial application scenario such as Xiaomi's recommender system. In this paper, we propose a new matrix factorization inspired by the theory of power law and GloVe. Instead of the exponential nature of GloVe model, we take advantage of Pareto Distribution to model our loss function. Our method is explainable in theory and easy-to-implement in practice. In the experiment section, we prove our approach is superior to vanilla matrix factorization technique and comparable with GloVe-based model in both accuracy and fairness metrics.
Distributed word embeddings such as Word2Vec and GloVe have been widely adopted in industrial context settings. Major technical applications of GloVe include recommender systems and natural language processing. The fundamental theory behind GloVe relies on the selection of a weighting function in the weighted least squres formulation that computes the powered ratio of word occurrence count and the maximum word count in the corpus. However, the initial formulation of GloVe is not theoretically sound in two aspects, namely the selection of the weighting function and its power exponent is ad-hoc. In this paper, we utilize the theory of extreme value analysis and propose a theoretically accurate version of GloVe. By reformulating the weighted least squares loss function as the expected loss function and accurately choosing the power exponent, we create a theoretically accurate version of GloVe. We demonstrate the competitiveness of our algorithm and show that the initial formulation of GloVe with the suggested optimal parameter can be viewed as a special case of our paradigm.
Movie Recommender System is widely applied in commercial environments such as NetFlix and Tubi. Classic recommender models utilize technologies such as collaborative filtering, learning to rank, matrix factorization and deep learning models to achieve lower marketing expenses and higher revenues. However, audience of movies have different ratings of the same movie in different contexts. Important movie watching contexts include audience mood, location, weather, etc. Tobe able to take advantage of contextual information is of great benefit to recommender builders. However, popular techniques such as tensor factorization consumes an impractical amount of storage, which greatly reduces its feasibility in real world environment. In this paper, we take advantage of the MatMat framework, which factorizes matrices by matrix fitting to build a context-aware movie recommender system that is superior to classic matrix factorization and comparable in the fairness metric.