Federated learning (FL) is a machine learning setting where many clients (e.g. mobile devices or whole organizations) collaboratively train a model under the orchestration of a central server (e.g. service provider), while keeping the training data decentralized. FL embodies the principles of focused data collection and minimization, and can mitigate many of the systemic privacy risks and costs resulting from traditional, centralized machine learning and data science approaches. Motivated by the explosive growth in FL research, this paper discusses recent advances and presents an extensive collection of open problems and challenges.

High-mobility group box-1 (HMGB1) plays a context-dependent role in autophagy, which is required for hepatic stellate cells (HSCs) activation. However, the significance of HMGB1-induced HSCs autophagy in liver fibrosis has not been elucidated. Here, we first documented an enrichment of peripheral and intrahepatic HMGB1 signal in hepatitis B virus (HBV)-related liver fibrosis progression, and presented a direct evidence of anatomic proximity of HMGB1 with a-SMA (a marker for HSCs activation) in cirrhotic liver specimens. Then, we demonstrated the autophagy-inducing effects by serum-sourced HMGB1 in both primary murine HSCs and human HSCs cell line (LX-2), reflected by increased number of autophagic vacuoles (AVs) under the transmission electron microscope (TEM) and up-regulated protein expression of lipidated microtubule-associated light chain 3 (LC3-II) (a marker for autophagosome) in Western blot analysis. Intriguingly, there is a possible translocation of endogenous HMGB1 from the nucleus to cytoplasm to extracellular space, during exogenous HMGB1-induced HSCs autophagy. Meanwhile, the dose- and time-dependent effects by recombinant HMGB1 (rHMGB1) in enhancing LX-2 autophagy and fibrogenesis have been revealed with activated extracellular regulated protein kinase (ERK)/c-Jun N-terminal kinase (JNK) mitogen-activated protein kinase (MAPK) and restrained mammalian target of rapamycin (mTOR)/STAT3 signaling pathways. Additionally, the ERK or JNK inhibitor could not only inhibit rHMGB1-induced autophagy and fibrogenesis in LX-2 cells, but also restore the suppressed mTOR and STAT3 pathways. Furthermore, using LC3-siRNA transfected LX-2, we found HMGB1-induced fibrogenesis is dependent on its autophagy-inducing effects. Finally, we elucidated the involvement of extracellular HMGB1-receptor for advenced glycation end product (RAGE) axis and endogenous HMGB1 in exogenous HMGB1-induced effects. Our findings could open new perspectives in developing an antifibrotic therapy by targetting the HSCs autophagy.

International audienceLarge-scale evaporation experiments were conducted on bare sand using an environmental chamber. Four different atmospheric conditions and various drying durations were imposed to soil sample. Both the atmospheric parameters (air flow rate, relative humidity and temperature) and the response of soil (volumetric water content, temperature and soil suction) were monitored simultaneously. Notably, the temperature and matric suction at soil surface were monitored using infrared thermometer and high-capacity tensiometer, respectively. The results show that the air and soil temperatures depend on the evaporation process and atmospheric conditions. In addition, volumetric water content in the near-surface zone is strongly affected by the evaporation process and changes linearly over depth. The evaporation rate is strongly dependent on the air conditions