The k-means clustering algorithm is considered one of the most powerful and popular data mining algorithms in the research community. However, despite its popularity, the algorithm has certain limitations, including problems associated with random initialization of the centroids which leads to unexpected convergence. Additionally, such a clustering algorithm requires the number of clusters to be defined beforehand, which is responsible for different cluster shapes and outlier effects. A fundamental problem of the k-means algorithm is its inability to handle various data types. This paper provides a structured and synoptic overview of research conducted on the k-means algorithm to overcome such shortcomings. Variants of the k-means algorithms including their recent developments are discussed, where their effectiveness is investigated based on the experimental analysis of a variety of datasets. The detailed experimental analysis along with a thorough comparison among different k-means clustering algorithms differentiates our work compared to other existing survey papers. Furthermore, it outlines a clear and thorough understanding of the k-means algorithm along with its different research directions.

In recent years, preference-based human feedback mechanisms have become essential for enhancing model performance across diverse applications, including conversational AI systems such as ChatGPT. However, existing approaches often neglect critical aspects, such as model uncertainty and the variability in feedback quality. To address these challenges, we introduce an entropy-based human feedback framework for contextual bandits, which dynamically balances exploration and exploitation by soliciting expert feedback only when model entropy exceeds a predefined threshold. Our method is model-agnostic and can be seamlessly integrated with any contextual bandit agent employing stochastic policies. Through comprehensive experiments, we show that our approach achieves significant performance improvements while requiring minimal human feedback, even under conditions of suboptimal feedback quality. This work not only presents a novel strategy for feedback solicitation but also highlights the robustness and efficacy of incorporating human guidance into machine learning systems. Our code is publicly available: https://github.com/BorealisAI/CBHF

We study the problem of off-policy critic evaluation in several variants of value-based off-policy actor-critic algorithms. Off-policy actor-critic algorithms require an off-policy critic evaluation step, to estimate the value of the new policy after every policy gradient update. Despite enormous success of off-policy policy gradients on control tasks, existing general methods suffer from high variance and instability, partly because the policy improvement depends on gradient of the estimated value function. In this work, we present a new way of off-policy policy evaluation in actor-critic, based on the doubly robust estimators. We extend the doubly robust estimator from off-policy policy evaluation (OPE) to actor-critic algorithms that consist of a reward estimator performance model. We find that doubly robust estimation of the critic can significantly improve performance in continuous control tasks. Furthermore, in cases where the reward function is stochastic that can lead to high variance, doubly robust critic estimation can improve performance under corrupted, stochastic reward signals, indicating its usefulness for robust and safe reinforcement learning.

The policy gradient theorem is defined based on an objective with respect to the initial distribution over states. In the discounted case, this results in policies that are optimal for one distribution over initial states, but may not be uniformly optimal for others, no matter where the agent starts from. Furthermore, to obtain unbiased gradient estimates, the starting point of the policy gradient estimator requires sampling states from a normalized discounted weighting of states. However, the difficulty of estimating the normalized discounted weighting of states, or the stationary state distribution, is quite well-known. Additionally, the large sample complexity of policy gradient methods is often attributed to insufficient exploration, and to remedy this, it is often assumed that the restart distribution provides sufficient exploration in these algorithms. In this work, we propose exploration in policy gradient methods based on maximizing entropy of the discounted future state distribution. The key contribution of our work includes providing a practically feasible algorithm to estimate the normalized discounted weighting of states, i.e, the \textit{discounted future state distribution}. We propose that exploration can be achieved by entropy regularization with the discounted state distribution in policy gradients, where a metric for maximal coverage of the state space can be based on the entropy of the induced state distribution. The proposed approach can be considered as a three time-scale algorithm and under some mild technical conditions, we prove its convergence to a locally optimal policy. Experimentally, we demonstrate usefulness of regularization with the discounted future state distribution in terms of increased state space coverage and faster learning on a range of complex tasks.

We propose a theoretical framework for approximate planning and learning in partially observed systems. Our framework is based on the fundamental notion of information state. We provide two equivalent definitions of information state -- i) a function of history which is sufficient to compute the expected reward and predict its next value; ii) equivalently, a function of the history which can be recursively updated and is sufficient to compute the expected reward and predict the next observation. An information state always leads to a dynamic programming decomposition. Our key result is to show that if a function of the history (called approximate information state (AIS)) approximately satisfies the properties of the information state, then there is a corresponding approximate dynamic program. We show that the policy computed using this is approximately optimal with bounded loss of optimality. We show that several approximations in state, observation and action spaces in literature can be viewed as instances of AIS. In some of these cases, we obtain tighter bounds. A salient feature of AIS is that it can be learnt from data. We present AIS based multi-time scale policy gradient algorithms. and detailed numerical experiments with low, moderate and high dimensional environments.