Constructing narrative using a generative model and continuous action policies
نویسندگان
چکیده
This paper proposes a method for learning how to generate narrative by recombining sentences from a previous collection. Given a corpus of story events categorised into 9 topics, we approximate a deep reinforcement learning agent policy to recombine them in order to satisfy narrative structure. We also propose an evaluation of such a system. The evaluation is based on coherence, interest, and topic, in order to figure how much sense the generated stories make, how interesting they are, and examine whether new narrative topics can emerge.
منابع مشابه
End-to-End Differentiable Adversarial Imitation Learning
Generative Adversarial Networks (GANs) have been successfully applied to the problem of policy imitation in a model-free setup. However, the computation graph of GANs, that include a stochastic policy as the generative model, is no longer differentiable end-to-end, which requires the use of high-variance gradient estimation. In this paper, we introduce the Modelbased Generative Adversarial Imit...
متن کاملA Novel Continuous KNN Prediction Algorithm to Improve Manufacturing Policies in a VMI Supply Chain
This paper examines and compares various manufacturing policies which manufacturer may adopt so as to improve the performance of a vendor managed inventory (VMI) partnership. The goal is to maximize the combined cumulative profit of supply chain while minimizing relevant inventory management costs. The supply chain is a two-level system with a single manufacturer and single retailer at each lev...
متن کاملThe Actor-Topic Model for Extracting Social Networks in Literary Narrative
We present a generative model for conversational dialogues, namely the actortopic model (ACTM), that extend the author-topic model (Rosen-Zvi, et.al, 2004) to identify actors of given conversation in literary narratives. Thus ACTM assigns each instance of quoted speech to an appropriate character. We model dialogues in a literary text, which take place between two or more actors conversing on d...
متن کاملManifold Embeddings for Model-Based Reinforcement Learning of Neurostimulation Policies
Real-world reinforcement learning problems often exhibit nonlinear, continuous-valued, noisy, partially-observable state-spaces that are prohibitively expensive to explore. The formal reinforcement learning framework, unfortunately, has not been successfully demonstrated in a real-world domain having all of these constraints. We approach this domain with a two-part solution. First, we overcome ...
متن کاملConstructing Stories in a Foreign Language: Analysis of Iranian EFL Learners’ Lived Narratives Structure
Most popular models of narratives and narrative analyses have been drawn on native stories, yet EFL learners’ narratives have not received due narrative analysis. The present study then aims at scrutinizing the structure of personal English stories as told by EFL learners. To this aim, three hundred narratives were collected through classroom discussions and interviews. Qualitative analysis met...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2017