Deep reinforcement learning (DRL) has been envisioned to have a competitive edge in quantitative finance. However, there is a steep development curve for quantitative traders to obtain an agent that automatically positions to win in the market, namely to decide where to trade, at what price and what quantity, due to the error-prone programming and arduous debugging. In this paper, we present the first open-source framework FinRL as a full pipeline to help quantitative traders overcome the steep learning curve. FinRL is featured with simplicity, applicability and extensibility under the key principles, full-stack framework, customization, reproducibility and hands-on tutoring.Embodied as a three-layer architecture with modular structures, FinRL implements fine-tuned state-of-the-art DRL algorithms and common reward functions, while alleviating the debugging workloads. Thus, we help users pipeline the strategy design at a high turnover rate. At multiple levels of time granularity, FinRL simulates various markets as training environments using historical data and live trading APIs. Being highly extensible, FinRL reserves a set of user-import interfaces and incorporates trading constraints such as market friction, market liquidity and investor's risk-aversion. Moreover, serving as practitioners' stepping stones, typical trading tasks are provided as step-by-step tutorials, e.g., stock trading, portfolio allocation, cryptocurrency trading, etc. CCS CONCEPTS• Computing methodologies → Machine learning; Markov decision processes; Reinforcement learning.

Purpose New opportunities to nurture good ideas for innovation arise as firms use web-based ideation platforms for collective idea generation and development. What influences creative performance in firm-internal collective idea development is however not as well researched as idea generation and thus an important area of research is the feedback and commenting on ideas. More specifically, the purpose of this paper is to explore the role of feedback timeliness and knowledge overlap between feedback providers and ideas in collective firm-internal online idea development. Design/methodology/approach An empirical study has been performed, drawing on data collected from a Swedish multi-national company using a web-based system for collective firm-internal ideation. The investigation explicitly captures the effects on ideation performance played by idea development contributions, in terms of feedback timeliness and knowledge overlap between feedback providers and ideas. Findings The empirical results show that idea development is significantly influenced by feedback timeliness as well as by the knowledge overlap between feedback providers and ideas. Specifically, it is found that longer time to feedback and an increased knowledge overlap result in an increased likelihood of idea acceptance. However, beyond a certain point, the positive effects of a longer time to feedback and increased knowledge overlap decrease, resulting in curvilinear relationships with idea acceptance. Research limitations/implications The results do not only shed new light on theory about collective idea development, but also provides management implications for collective firm-internal ideation. As the data used in the study has been collected in one single firm, care should be taken in generalizing the results to other domains. Practical implications The results inform managers that it is not always better to involve more individuals in these emergent and distributed ideation systems, but that it might be beneficial to take measures to exercise some control in terms of when distributed and diverse employees can freely join in and out, especially considering the diversity of ideas, comments and creators. Originality/value The results from the empirical study reveal the effects of feedback timeliness and knowledge overlap on idea development. This provides us with new insights on the complex dynamics at place in collective firm-internal idea development and offers implications for how we can fruitfully manage this process.

PurposeDespite the fact that user participation (UP) has been highlighted as an important aspect in innovation, previous findings on its relationship with service innovation performance (SIP) are inconsistent. This study aims to investigate the relationships among UP, knowledge management capability (KMC) and SIP, especially in the digital age, inspired by the theories of knowledge-based and absorptive capacity.Design/methodology/approachBased on a sample of 252 Chinese e-commerce enterprises, this study adopts a hierarchical regression analysis and bootstrap method to test the theoretical framework and research hypotheses.FindingsUP and KMC have positive effects on SIP, respectively. KMC plays a mediating role in the effect of UP on SIP. Furthermore, the intermediary role of KMC varies in different sub-paths between UP and SIP.Originality/valueFirst, this study provides some explanations for inconsistent arguments on the relationship between UP and service innovation. Second, with the consideration of specific dimensions of UP and SIP, the mediation role of KMC varies in different sub-paths has been recognized, which provides a deeper understanding of the relationship between UP and SIP. Third, this study opens the discussion about how to realize SIP more effectively in the digital age, advancing theoretical and practical developments on service innovation.

No abstract

Solvers' participation is essential for successful implementation of crowdsourcing contests for problem solving (CCPS). Many efforts have been made to investigate solvers' various participation behaviours in CCPS. Whether or not a solver will conduct a behaviour is the result of decision making. However, to our knowledge, few studies concentrated on solvers' participation from a decision process perspective and little is known about the factors that influence the decisions that solvers are likely to make. This study aims to develop a framework for demonstrating solvers' decisions and their relations, thereafter identify the factors that affect each of decision makings. It does so through the qualitative structured interviews conducted with solvers in a crowdsourcing platform. The interviews capture four major interrelated solvers' decisions that are decisions of participation in CCPS, platform selection, contest selection and determination of effort level, respectively. Moreover, the factors including solvers' motives, solvers' individual characteristics and incentives, and their roles in each of solvers' decision makings are presented. The findings improve the understanding of solvers' participation in CCPS from a decision process perspective. With the further comprehending of factors that affect solvers' decision makings, this study provides practical implications for crowdsourcing platforms to improve their services for solvers.