Kshitij Jerath

Hossein Haeri, Reza Ahmadzadeh, Kshitij Jerath

In this work, we integrate `social' interactions into the MARL setup through a user-defined relational network and examine the effects of agent-agent relations on the rise of emergent behaviors. Leveraging insights from sociology and neuroscience, our proposed framework models agent relationships using the notion of Reward-Sharing Relational Networks (RSRN), where network edge weights act as a measure of how much one agent is invested in the success of (or `cares about') another. We construct relational rewards as a function of the RSRN interaction weights to collectively train the multi-agent system via a multi-agent reinforcement learning algorithm. The performance of the system is tested for a 3-agent scenario with different relational network structures (e.g., self-interested, communitarian, and authoritarian networks). Our results indicate that reward-sharing relational networks can significantly influence learned behaviors. We posit that RSRN can act as a framework where different relational networks produce distinct emergent behaviors, often analogous to the intuited sociological understanding of such networks.

Kshitij Kumar Srivastava, Kshitij Jerath

Hierarchical Reinforcement Learning (HRL) intends to separate strategic planning from primitive execution. It has been widely successful in solving long-horizon and complex tasks, where flat-RL algorithms have difficulty in learning. However, while the low-level agent in HRL benefits from dense feedback and abundant trial opportunities, the high-level agent receives sparse, delayed feedback from the environment and its performance depends on the low-level execution capability. In this paper, we study whether subgoal selection by the high-level agent can be performed more strategically, by providing it with dynamics-aware intrinsic motivation. Since motivation based on primitive transition dynamics would require broad coverage of the state-action space, we propose to use coarse dynamics, i.e., environment transitions aggregated over multiple steps at the temporal scale at which the high-level agent operates. This approach stabilizes the high-level policy by learning to minimize the predictive uncertainty associated with the coarse dynamics, and provides a guided structure for navigation. We model the predictive uncertainty by evaluating different dispersion metrics as approximated by a Mixture Density Network (MDN). Empirically, we observe that a dense, dynamics-aware intrinsic reward leads to risk-averse subgoal selection, enabling it to outperform state-of-the-art HRL methods in non-stationary long-horizon environments.

Zhaohui Yang, Kshitij Jerath

Recent endeavors aimed at forecasting future traffic flow states through deep learning encounter various challenges and yield diverse outcomes. A notable obstacle arises from the substantial data requirements of deep learning models, a resource often scarce in traffic flow systems. Despite the abundance of domain knowledge concerning traffic flow dynamics, prevailing deep learning methodologies frequently fail to fully exploit it. To address these issues, we propose an innovative solution that merges Convolutional Neural Networks (CNNs) with Long Short-Term Memory (LSTM) architecture to enhance the prediction of traffic flow dynamics. A key revelation of our research is the feasibility of sampling training data for large traffic systems from simulations conducted on smaller traffic systems. This insight suggests the potential for referencing a macroscopic-level distribution to inform the sampling of microscopic data. Such sampling is facilitated by the observed scale invariance in the normalized energy distribution of the statistical mechanics model, thereby streamlining the data generation process for large-scale traffic systems. Our simulations demonstrate promising agreement between predicted and actual traffic flow dynamics, underscoring the efficacy of our proposed approach.

Niket Kathiriya, Hossein Haeri, Cindy Chen et al.

Many modern systems, such as financial, transportation, and telecommunications systems, are time-sensitive in the sense that they demand low-latency predictions for real-time decision-making. Such systems often have to contend with continuous unbounded data streams as well as concept drift, which are challenging requirements that traditional regression techniques are unable to cater to. There exists a need to create novel data stream regression methods that can handle these scenarios. We present a database-inspired datastream regression model that (a) uses inspiration from R*-trees to create granules from incoming datastreams such that relevant information is retained, (b) iteratively forgets granules whose information is deemed to be outdated, thus maintaining a list of only recent, relevant granules, and (c) uses the recent data and granules to provide low-latency predictions. The R*-tree-inspired approach also makes the algorithm amenable to integration with database systems. Our experiments demonstrate that the ability of this method to discard data produces a significant order-of-magnitude improvement in latency and training time when evaluated against the most accurate state-of-the-art algorithms, while the R*-tree-inspired granulation technique provides competitively accurate predictions