Sökning: "Ablationsstudie"

Hittade 5 uppsatser innehållade ordet Ablationsstudie.

1. 1. Analyzing How Blended Emotions are Expressed using Machine Learning Methods

   Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Disen Ling; [2023]
   Nyckelord :Blended emotion; Supervised learning; Model generalization capability; Ablation study; Blandade känslor; Övervakad inlärning; Modellens generaliseringsförmåga; Ablationsstudie;

   Sammanfattning : Blended emotion is a classification of emotional experiences that involve the combination of multiple emotions. Research on the expression of blended emotions allows researchers to understand how different emotions interact and coexist in an individual’s emotional experience. LÄS MER

3. 2. Model-based Residual Policy Learning for Sample Efficient Mobile Network Optimization

   Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Viktor Eriksson Möllerstedt; [2022]
   Nyckelord :Reinforcement Learning; Sample Efficiency; Model-based; Expert Policy; Remote Electrical Tilt; Telecommunication; Förstärkande inlärning; dataeffektivitet; modell-baserad; expert-policy; fjärrstyrning av antenners nedåtlutning; telekommunikation;

   Sammanfattning : Reinforcement learning is a powerful tool which enables an agent to learn how to control complex systems. However, during the early phases of training, the performance is often poor. LÄS MER

4. 3. Evaluating behaviour tree integration in the option critic framework in Starcraft 2 mini-games with training restricted by consumer level hardware

   Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Fredrik Lundberg; [2022]
   Nyckelord :Hierarchical reinforcement learning; reinforcement learning; behaviour tree; option critic; Starcraft 2; Hierarkiskt förstärkningsinlärning; förstärkningsinlärning; beteendeträd; option critic; Starcraft 2;

   Sammanfattning : This thesis investigates the performance of the option critic (OC) framework combined with behaviour trees (BTs) in Starcraft 2 mini-games when training time is constrained by a time frame limited by consumer level hardware. We test two such combination models: BTs as macro actions (OCBT) and BTs as options (OCBToptions) and measure the relative performance to the plain OC model through an ablation study. LÄS MER

5. 4. Using Bidirectional Encoder Representations from Transformers for Conversational Machine Comprehension

   Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Evangelina Gogoulou; [2019]
   Nyckelord :conversational machine comprehension; question answering; transformers; self-attention; language modelling; samtalsmaskinförståelse; frågesvar; transformatorer; självuppmärksamhet; språkmodellering;

   Sammanfattning : Bidirectional Encoder Representations from Transformers (BERT) is a recently proposed language representation model, designed to pre-train deep bidirectional representations, with the goal of extracting context-sensitive features from an input text [1]. One of the challenging problems in the field of Natural Language Processing is Conversational Machine Comprehension (CMC). LÄS MER

7. 5. Ablation Programming for Machine Learning

   Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Sina Sheikholeslami; [2019]
   Nyckelord :Distributed Machine Learning; Distributed Systems; Ablation Studies; Apache Spark; Keras; Hopsworks;

   Sammanfattning : As machine learning systems are being used in an increasing number of applications from analysis of satellite sensory data and health-care analytics to smart virtual assistants and self-driving cars they are also becoming more and more complex. This means that more time and computing resources are needed in order to train the models and the number of design choices and hyperparameters will increase as well. LÄS MER