Actor-Critic Reinforcement Learning and Application in Developing Computer-Vision-Based Interface Tracking

This paper synchronizes control theory with computer vision by formalizing object tracking as a sequential decision-making process.A reinforcement learning(RL)agent successfully tracks an interface between two liquids,which is often a critical variable to track in many chemical,petrochemical,metallurgical,and oil industries.This method utilizes less than 100 images for creating an environment,from which the agent generates its own data without the need for expert knowledge.Unlike supervised learning(SL)methods that rely on a huge number of parameters,this approach requires far fewer parameters,which naturally reduces its maintenance cost.Besides its frugal nature,the agent is robust to environmental uncertainties such as occlusion,intensity changes,and excessive noise.From a closed-loop control context,an interface location-based deviation is chosen as the optimization goal during training.The methodology showcases RL for real-time object-tracking applications in the oil sands industry.Along with a presentation of the interface tracking problem,this paper provides a detailed review of one of the most effective RL methodologies:actor–critic policy.

Projective Blaschke Manifolds

In this paper we define the concept of projective Blaschke manifolds and extend the theory of equiaffine differential geometry to the projective Blaschke manifolds.

Spatio-temporal dynamics of phytohormones in the tomato graft healing process

Graft healing involves a series of cytological and molecular events including wound responses, callus formation and vascular bundle remodelling. Hormones are important signalling molecules regulating plant development and responses to environmental stimuli. However,the detailed dynamics of phytohormones in graft healing remain elusive. In this research, internodes above and below the graft site were harvested from 0 to 168 h after grafting(HAG), and liquid chromatography tandem mass spectrometry(LC-MS/MS) was used to determinate jasmonic acid, auxin, cytokinin, ethylene, salicylic acid, abscisic acid and gibberellin levels during the graft healing process. Uniform manifold approximation and projection(UMAP) and k-means analyses were performed to explore hormone spatio-temporal dynamics. We found the stage-specific and asymmetric accumulation of phytohormones in the tomato graft healing process. At the early healing stage(before vascular bundle reconnection), IAA, cZ, ABA, JA and SA mainly accumulated above the graft site, while tZ and ACC mainly accumulated below the graft site. MEIAA, ICAld and IP mainly accumulated at the later stage. Comminated with the healing process, we suggested that JA is mainly involved in wound responses, IAA is beneficial to the formation of callus and vascular cell development, tZ promotes cell division, and IP is linked to vascular bundle remodelling. In addition, expression of JA-related genes SlMYC2 and SlJAZ2, IAA-related gene SlIAA1, tZ-related genes SlHP2 and SlRR8, and IP-related gene SlRR9 correlated with hormone accumulation. The findings provide important information about the hormones and genes involved in the tomato graft healing process.