A Hybrid Sparse Representation Model for Image Restoration

Group-based sparse representation (GSR) uses image nonlocal self-similarity (NSS) prior to grouping similar image patches, and then performs sparse representation. However, the traditional GSR model restores the image by training degraded images, which leads to the inevitable over-fitting of the data in the training model, resulting in poor image restoration results. In this paper, we propose a new hybrid sparse representation model (HSR) for image restoration. The proposed HSR model is improved in two aspects. On the one hand, the proposed HSR model exploits the NSS priors of both degraded images and external image datasets, making the model complementary in feature space and the plane. On the other hand, we introduce a joint sparse representation model to make better use of local sparsity and NSS characteristics of the images. This joint model integrates the patch-based sparse representation (PSR) model and GSR model, while retaining the advantages of the GSR model and the PSR model, so that the sparse representation model is unified. Extensive experimental results show that the proposed hybrid model outperforms several existing image recovery algorithms in both objective and subjective evaluations.

Deep Learning for Alzheimer’s Disease Drug Repurposing using Knowledge Graph and Multi-level Evidence

ABSTRACTDeveloping drugs for treating Alzheimer’s disease (AD) has been extremely challenging and costly due to limited knowledge on underlying biological mechanisms and therapeutic targets. Repurposing drugs or their combination has shown potential in accelerating drug development due to the reduced drug toxicity while targeting multiple pathologies. To address the challenge in AD drug development, we developed a multi-task machine learning pipeline to integrate a comprehensive knowledge graph on biological/pharmacological interactions and multi-level evidence on drug efficacy, to identify repurposable drugs and their combination candidates. We developed and computationally validated a heterogeneous graph representation model with transfer learning from universal biomedical databases and with joint optimization with AD risk genes. Using the drug embedding from the heterogeneous graph representation model, we ranked drug candidates based on evidence from post-treatment transcriptomic patterns, mechanistic efficacy in preclinical models, population-based treatment effect, and Phase II/III clinical trials. We experimentally validated the top-ranked candidates in neuronal cells, identifying drug combinations with efficacy in reducing oxidative stress and safety in maintaining neuronal viability and morphology. Our neuronal response experiments confirmed several biologically efficacious drug combinations (e.g., Galantamine + Mifepristone). This pipeline showed that harmonizing heterogeneous and complementary data/knowledge, including human interactome, transcriptome patterns, experimental efficacy, and real-world patient data shed light on the drug development of complex diseases.

Efficient Estimate of Low-Frequency Words’ Embeddings Based on the Dictionary: A Case Study on Chinese

Obtaining high-quality embeddings of out-of-vocabularies (OOVs) and low-frequency words is a challenge in natural language processing (NLP). To efficiently estimate the embeddings of OOVs and low-frequency words, we propose a new method that uses the dictionary to estimate the embeddings of OOVs and low-frequency words. More specifically, the explanatory note of an entry in dictionaries accurately describes the semantics of the corresponding word. Naturally, we adopt the sentence representation model to extract the semantics of the explanatory note and regard the semantics as the embedding of the corresponding word. We design a new sentence representation model to encode sentences to extract the semantics from the explanatory notes of entries more efficiently. Based on the assumption that the higher quality of word embeddings will lead to better performance, we design an extrinsic experiment to evaluate the quality of low-frequency words’ embeddings. The experimental results show that the embeddings of low-frequency words estimated by our proposed method have higher quality. In addition, both intrinsic and extrinsic experiments show that our proposed sentence representation model can represent the semantics of sentences well.

Context-Based Movie Searching Using Semantic Web

When searching for a movie, users often remember only the incidents happened in the movie instead of the actors or directors of that movie. However, these searches are not supported in our current movie information systems as data query is usually based on keywords. This research proposes a solution to search and query movies based on the content of the movie or the incidents happened in the movie. In our research, we have analysed and designed some movie representation models suitable for context-based movie searching. We also propose a quadruple-based representation model to resolve the disadvantages in current semantic web’s triple-based model. Our system is capable of processing user requests precisely and has proven to have advantages over current movie information systems.