Deep Reinforcement Learning has achieved superhuman performance in many tasks, such as robotic control or autonomous driving. Algorithms in Deep Reinforcement Learning still suffer from a sample efficiency problem, where, in many cases, millions of samples are needed to achieve good performance. Recently, Bayesian uncertainty-based algorithms have gained traction. This work focuses on providing a better understanding of the behaviour of Langevin Monte Carlo algorithms for Bayesian posterior approximation applied on top of Q-learning. This research builds on top of already existing algorithms, aiming to provide a better understanding of the underlying mechanics that drive them. We provide empirical experimentation with different hyperparameters in three different environments. Our results suggest that hyperparameters that were previously thought not to have a big impact on the algorithms are crucial for deep exploration.

Efficient exploration is a major issue in reinforcement learning, particularly in environments with sparse rewards. In these environments, traditional methods like e-greedy fail to efficiently reach an optimal policy. A new method proposed by Fortunato, et al. Fortunato, et al. showed promise by improving efficiency on RL tasks such as Atari games, by driving exploration with learned perturbations of the network weights. Three different types of settings were investigated in order to test the robustness of a contextual bandit implemented with this proposed method: 1. ContextualBandit-v2; a bandit with multiple predefined functions mapping the 1-dimensional continuous input to the reward, 2. MNISTBandit-v0; a bandit rewarding correct identification of MNIST dataset images, and 3. NNBandit-v0; a bandit with the reward being determined by a neural network. Furthermore, non-stationary variants of environment 1. and 3. were tested. A slight variation in hyperparameter sensitivity between environments was observed and a generally optimal set was determined. Overall, NoisyNet-DQNs (Deep Q-Networks) achieved performance comparable to regular DQNs, though often slightly lower. In the high-dimensional stationary MNISTBandit-v0 environment, NoisyNet-DQN converged to an optimal policy slightly faster, at the cost of a larger variation in performance.

We present a large-scale empirical study of Bootstrapped DQN (BDQN) and Randomized-Prior BDQN (RP-BDQN) in the DeepSea environment, aimed at characterizing their scaling properties. Our primary contribution is a unified scaling law that accurately models the probability of reward discovery as a function of task hardness and ensemble size. This law is parameterized by a method-dependent effectiveness factor, $\psi$.Under this framework, RP-BDQN demonstrates substantially higher effectiveness ($\psi \approx 0.87$) compared to BDQN ($\psi \approx 0.80$), enabling it to solve more challenging tasks.Our analysis reveals that this advantage stems from RP-BDQN's sustained ensemble diversity, which mitigates the posterior collapse observed in BDQN.Furthermore, we demonstrate diminishing returns in performance for ensemble sizes $K>10$. These results offer practical guidance for ensemble configuration and raise new theoretical questions surrounding the effectiveness parameter $\psi$.

This paper investigates how Random Network Distillation (RND), coupled with Boltzmann exploration, influences exploration behaviour and learning dynamics in value-based agents such as Deep Q-Learning (DQN) across a range of environments, from classic control tasks to behaviour suite benchmarks and contextual bandits. The study addresses the sensitivity of RND to key hyperparameters, the impact of exploration strategy design, and the transferability of settings across tasks. The results reveal that RND remains benefitial within DQN in both sequential and non-sequential tasks, but requires careful tuning of reward scaling, temperature, and network capacity to be effective. No universal hyperparameter configuration generalizes across environments, and inappropriate tuning can lead to unstable learning or suboptimal outcomes. These findings provide practical insights into the strengths and limitations of applying RND within value-based reinforcement learning frameworks.