Progress in Neural NLP:Modeling,Learning,and Reasoning

Natural language processing(NLP)is a subfield of artificial intelligence that focuses on enabling computers to understand and process human languages.In the last five years,we have witnessed the rapid development of NLP in tasks such as machine translation,question-answering,and machine reading comprehension based on deep learning and an enormous volume of annotated and unannotated data.In this paper,we will review the latest progress in the neural network-based NLP framework(neural NLP)from three perspectives:modeling,learning,and reasoning.In the modeling section,we will describe several fundamental neural network-based modeling paradigms,such as word embedding,sentence embedding,and sequence-to-sequence modeling,which are widely used in modern NLP engines.In the learning section,we will introduce widely used learning methods for NLP models,including supervised,semi-supervised,and unsupervised learning;multitask learning;transfer learning;and active learning.We view reasoning as a new and exciting direction for neural NLP,but it has yet to be well addressed.In the reasoning section,we will review reasoning mechanisms,including the knowledge,existing non-neural inference methods,and new neural inference methods.We emphasize the importance of reasoning in this paper because it is important for building interpretable and knowledgedriven neural NLP models to handle complex tasks.At the end of this paper,we will briefly outline our thoughts on the future directions of neural NLP.

Deep Reinforcement Learning Enabled Bi-level Robust Parameter Optimization of Hydropower-dominated Systems for Damping Ultra-low Frequency Oscillation

This paper proposes a robust and computationally efficient control method for damping ultra-low frequency oscillations(ULFOs) in hydropower-dominated systems. Unlike the existing robust optimization based control formulation that can only deal with a limited number of operating conditions, the proposed method reformulates the control problem into a bi-level robust parameter optimization model. This allows us to consider a wide range of system operating conditions. To speed up the bi-level optimization process, the deep deterministic policy gradient(DDPG) based deep reinforcement learning algorithm is developed to train an intelligent agent. This agent can provide very fast lower-level decision variables for the upper-level model, significantly enhancing its computational efficiency. Simulation results demonstrate that the proposed method can achieve much better damping control performance than other alternatives with slightly degraded dynamic response performance of the governor under various types of operating conditions.

Evolutionary engineering of Phaffia rhodozyma for astaxanthin-overproducing strain

Evolutionary engineering is a novel whole- genome wide engineering strategy inspired by natural evolution for strain improvement. Astaxanthin has been widely used in cosmetics, pharmaceutical and health care food due to its capability of quenching active oxygen. Strain improvement ofPhaffia rhodozyma, one of the main sources for natural astaxanthin, is of commercial interest for astaxanthin production. In this study a selection procedure was developed for adaptive evolution of P. rhodozyma strains under endogenetic selective pressure induced by additive in environmental niches. Six agents, which can induce active oxygen in cells, were added to the culture medium respectively to produce selective pressure in process of evolution. The initial strain, P. rhodozyma AS2-1557, was mutagenized to acquire the initial strain population, which was then cultivated for 550 h at selective pressure and the culture was transferred every 48h. Finally, six evolved strains were selected after 150 generations of evolution. The evolved strains produced up to 48.2% more astaxanthin than the initial strain. Our procedure may provide a promising alternative for improvement of highproduction strain.