In this project, we aim to reproduce previous resultsachieved with Deep Reinforcement Learning. We present theMarkov Decision Process model as well as the algorithms Q-learning and Deep Q-learning Network (DQN). We implement aDQN agent, first in an environment called CartPole, and later inthe game Pong.Our agent was able to solve the CartPole environment in lessthan 300 episodes. We assess the impact some of the parametershad on the agents performance. The performance of the agentis particularly sensitive to the learning rate and seeminglyproportional to the dimension of the neural network. The DQNagent implemented in Pong was unable to learn, performing atthe same level as an agent picking actions at random, despiteintroducing various modifications to the algorithm. We discusspossible sources of error, including the RAM used as input,possibly not containing sufficient information. Furthermore, wediscuss the possibility of needing additional modifications to thealgorithm in order to achieve convergence, as it is not guaranteedfor DQN. / Målet med detta projekt är att reproducera tidigare resultat som uppnåtts med Deep Reinforcement Learning. Vi presenterar Markov Decision Process-modellen samt algoritmerna Q-learning och Deep Q-learning Network (DQN). Vi implementerar en DQN agent, först i miljön CartPole, sedan i spelet Pong.  Vår agent lyckades lösa CartPole på mindre än 300 episoder. Vi gör en bedömning av vissa parametrars påverkan på agentens prestanda. Agentens prestanda är särskilt känslig för värdet på ”learning rate” och verkar vara proportionell mot dimensionen av det neurala nätverket. DQN-agenten som implementerades i Pong var oförmögen att lära sig och spelade på samma nivå som en agent som agerar slumpmässigt, trots att vi introducerade diverse modifikationer. Vi diskuterar möjliga felkällor, bland annat att RAM, som används som indata till agenten, eventuellt saknar tillräcklig information. Dessutom diskuterar vi att ytterligare modifikationer kan vara nödvändiga för uppnå konvergens eftersom detta inte är garanterat för DQN. / Kandidatexjobb i elektroteknik 2020, KTH, Stockholm

The purpose of this study was to better understand student learning in asynchronous computer-conference discourse (ASD) for non-native speakers of English in India through the Community of Inquiry (COI) framework. The study looked at ASD from an online course taught in the fall of 2015 to 25 students in a seminary in South India. All but one of the students were non-native speakers of English. The class consisted of 22 men and 3 women. Eight students spoke languages from the Dravidian family of languages (Malayalam, Tamil, Telegu and Kannada). Eight students were from the Northeastern states of Manipur, Nagaland and Tripura, where most languages are from the Sino-Tibetan family. Three students were native speakers of Indo-Aryan languages (Odiya and Assamese). Five students were from Myanmar representing several Sino-Tibetan languages. The COI is a framework used to understand learning in ASD, often used in online learning. To study the ASD of this group, critical discourse analysis (CDA) was used with the COI to capture the unique socio-cultural and linguistic conditions of this group. The study revealed that non-native speakers of English often reach the Exploration phase of learning but rarely show evidence of reaching the Resolution phase. This phenomenon was also observed in native English speakers as reported in the literature. Also, the structure of ASD showed that students took an examination approach to discussion shaped in part by their epistemology. This examination approach shaped how knowledge was constructed. CDA also showed that the discourse acquired an instructor-centered structure in which Resolution and Repair were initiated and finalized by the instructor. The study advances the COI framework by undergirding it with a theory of asynchronous discourse using critical discourse analysis and capturing cognitive, social and teaching presence phenomena for non-native speakers that were not observed through the traditional COI framework. These phenomena were driven by cultural, epistemological, and linguistic forces and require a rethinking of the COI for contexts outside of North America. The study also demonstrates that learning for non-native speakers in ASD is challenged by these very same forces. Therefore, design for online learning should account for these phenomena.

No description available.

No description available.