No abstract

Web-scale search systems typically tackle the scalability challenge with a two-step paradigm: retrieval and ranking. The retrieval step, also known as candidate selection, often involves extracting standardized entities, creating an inverted index, and performing term matching for retrieval. Such traditional methods require manual and time-consuming development of query models. In this paper, we discuss applying learning-to-retrieve technology to enhance LinkedIn's job search and recommendation systems. In the realm of promoted jobs, the key objective is to improve the quality of applicants, thereby delivering value to recruiter customers. To achieve this, we leverage confirmed hire data to construct a graph that evaluates a seeker's qualification for a job, and utilize learned links for retrieval. Our learned model is easy to explain, debug and adjust. On the other hand, the focus for organic jobs is to optimize seeker engagement. We accomplished this by training embeddings for personalized retrieval, fortified by a set of rules derived from the categorization of member feedbacks. In addition to a solution based on a conventional inverted index, we developed an on-GPU solution capable of supporting both KNN and term matching efficiently.

1. A clinical study to assess the interactions between albuvirtide (320 mg) and lopinavir/ritonavir (400/100 mg) was conducted in 10 HIV-1-infected subjects. Because albuvirtide requires a long period to achieve steady state, and extended monotherapy may lead to early resistance, it is unethical to take albuvirtide alone to achieve steady state. Therefore, a population pharmacokinetic model was developed to predict steady-state concentration-time curve of solely administered albuvirtide. 2. When albuvirtide and lopinavir/ritonavir were co-administered, the plasma concentration of albuvirtide when the infusion ended (C) increased by about 34%, but the geometric mean ratios and 90% confidence intervals (90% CIs) of AUC [1.09 (0.96-1.24)] and C [1.00 (0.83-1.20)] were within the range of 0.8-1.25. For lopinavir, the ratios (90% CIs) of AUC, C and C were 0.63 (0.49-0.82), 0.67 (0.53-0.86) and 0.65 (0.46-0.91); for ritonavir, those ratios (90% CIs) were 0.62 (0.42-0.91), 0.61 (0.38-0.99) and 0.72 (0.40-1.26), respectively. 3. Co-administration of albuvirtide with lopinavir/ritonavir has little effect on albuvirtide exposure, but it decreases the plasma exposures of lopinavir/ritonavir. However, the drug-drug interactions may not reduce the effectiveness of this co-therapy, the trough concentration of lopinavir may be sufficient and this combination could achieve similar clinical efficacy with marketed drugs. So, a phase 3 clinical trial without dose adjustment is underway to validate their effectiveness and safety.

No abstract

No abstract