Search Results for author:
Found 5 papers, 2 papers with code
Self-Supervised Contrastive BERT Fine-tuning for Fusion-based Reviewed-Item Retrieval
Experimental results show that Late Fusion contrastive learning for Neural RIR outperforms all other contrastive IR configurations, Neural IR, and sparse retrieval baselines, thus demonstrating the power of exploiting the two-level structure in Neural RIR approaches as well as the importance of preserving the nuance of individual review content via Late Fusion methods.
Efficient decorrelation of features using Gramian in Reinforcement Learning
Learning good representations is a long standing problem in reinforcement learning (RL).
Distributional Reinforcement Learning for Efficient Exploration
In distributional reinforcement learning (RL), the estimated distribution of value function models both the parametric and intrinsic uncertainties.
Deep Reinforcement Learning with Decorrelation
Further experiments on the losing games show that our decorelation algorithms can win over DQN and QR-DQN with a fined tuned regularization factor.
QUOTA: The Quantile Option Architecture for Reinforcement Learning
In this paper, we propose the Quantile Option Architecture (QUOTA) for exploration based on recent advances in distributional reinforcement learning (RL).
