Keshav Goyal

Keshav Goyal, Samuel Pearson, João Ribeiro et al.

We introduce and study the synthesis-sequencing channel, a two-stage model for DNA-based data storage that jointly captures synthesis and sequencing effects. The synthesis-sequencing channel provides a more nuanced and realistic model of the DNA storage process compared to prior work, as it distinguishes between physical coverage after synthesis and sequencing coverage after readout, relaxes the assumption of independent errors across reads, and naturally induces coverage bias through the composition of synthesis and sequencing stages. We establish the information-theoretic capacity of this channel by deriving matching converse and achievability bounds for the case where synthesis and sequencing errors are modeled by binary symmetric channels with possibly different error probabilities, under mild assumptions on the channel parameters. Our results reveal multiple trade-offs between physical coverage, synthesis errors, sequencing coverage, and sequencing errors that influence the maximum achievable rate for reliable data storage.

Sooraj Sathish, Keshav Goyal, Raghuram Bharadwaj Diddigi

Deep Reinforcement Learning (RL) has demonstrated success in solving complex sequential decision-making problems by integrating neural networks with the RL framework. However, training deep RL models poses several challenges, such as the need for extensive hyperparameter tuning and high computational costs. Transfer learning has emerged as a promising strategy to address these challenges by enabling the reuse of knowledge from previously learned tasks for new, related tasks. This avoids the need for retraining models entirely from scratch. A commonly used approach for transfer learning in RL is to leverage the internal representations learned by the neural network during training. Specifically, the activations from the last hidden layer can be viewed as refined state representations that encapsulate the essential features of the input. In this work, we investigate whether these representations can be used as input for training simpler models, such as linear function approximators, on new tasks. We observe that the representations learned by standard deep RL models can be highly correlated, which limits their effectiveness when used with linear function approximation. To mitigate this problem, we propose a novel deep Q-learning approach that introduces a regularization term to reduce positive correlations between feature representation of states. By leveraging these reduced correlated features, we enable more effective use of linear function approximation in transfer learning. Through experiments and ablation studies on standard RL benchmarks and MinAtar games, we demonstrate the efficacy of our approach in improving transfer learning performance and thereby reducing computational overhead.