T cells monitor the health status of cells by identifying foreign peptides displayed on their surface. T-cell receptors (TCRs), which are protein complexes found on the surface of T cells, are able to bind to these peptides. This process is known as TCR recognition and constitutes a key step for immune response. Optimizing TCR sequences for TCR recognition represents a fundamental step towards the development of personalized treatments to trigger immune responses killing cancerous or virus-infected cells. In this paper, we formulated the search for these optimized TCRs as a reinforcement learning (RL) problem, and presented a framework TCRPPO with a mutation policy using proximal policy optimization. TCRPPO mutates TCRs into effective ones that can recognize given peptides. TCRPPO leverages a reward function that combines the likelihoods of mutated sequences being valid TCRs measured by a new scoring function based on deep autoencoders, with the probabilities of mutated sequences recognizing peptides from a peptide-TCR interaction predictor. We compared TCRPPO with multiple baseline methods and demonstrated that TCRPPO significantly outperforms all the baseline methods to generate positive binding and valid TCRs. These results demonstrate the potential of TCRPPO for both precision immunotherapy and peptide-recognizing TCR motif discovery.

A unique challenge in Multi-Agent Reinforcement Learning (MARL) is the curse of multiagency, where the description length of the game as well as the complexity of many existing learning algorithms scale exponentially with the number of agents. While recent works successfully address this challenge under the model of tabular Markov Games, their mechanisms critically rely on the number of states being finite and small, and do not extend to practical scenarios with enormous state spaces where function approximation must be used to approximate value functions or policies. This paper presents the first line of MARL algorithms that provably resolve the curse of multiagency under function approximation. We design a new decentralized algorithm -- V-Learning with Policy Replay, which gives the first polynomial sample complexity results for learning approximate Coarse Correlated Equilibria (CCEs) of Markov Games under decentralized linear function approximation. Our algorithm always outputs Markov CCEs, and achieves an optimal rate of $\widetilde{\mathcal{O}}(\epsilon^{-2})$ for finding $\epsilon$-optimal solutions. Also, when restricted to the tabular case, our result improves over the current best decentralized result $\widetilde{\mathcal{O}}(\epsilon^{-3})$ for finding Markov CCEs. We further present an alternative algorithm -- Decentralized Optimistic Policy Mirror Descent, which finds policy-class-restricted CCEs using a polynomial number of samples. In exchange for learning a weaker version of CCEs, this algorithm applies to a wider range of problems under generic function approximation, such as linear quadratic games and MARL problems with low ''marginal'' Eluder dimension.

Inspired by the recent success of sequence modeling in RL and the use of masked language model for pre-training, we propose a masked model for pre-training in RL, RePreM (Representation Pre-training with Masked Model), which trains the encoder combined with transformer blocks to predict the masked states or actions in a trajectory. RePreM is simple but effective compared to existing representation pre-training methods in RL. It avoids algorithmic sophistication (such as data augmentation or estimating multiple models) with sequence modeling and generates a representation that captures long-term dynamics well. Empirically, we demonstrate the effectiveness of RePreM in various tasks, including dynamic prediction, transfer learning, and sample-efficient RL with both value-based and actor-critic methods. Moreover, we show that RePreM scales well with dataset size, dataset quality, and the scale of the encoder, which indicates its potential towards big RL models.

Knowing the learning dynamics of policy is significant to unveiling the mysteries of Reinforcement Learning (RL). It is especially crucial yet challenging to Deep RL, from which the remedies to notorious issues like sample inefficiency and learning instability could be obtained. In this paper, we study how the policy networks of typical DRL agents evolve during the learning process by empirically investigating several kinds of temporal change for each policy parameter. On typical MuJoCo and DeepMind Control Suite (DMC) benchmarks, we find common phenomena for TD3 and RAD agents: 1) the activity of policy network parameters is highly asymmetric and policy networks advance monotonically along very few major parameter directions; 2) severe detours occur in parameter update and harmonic-like changes are observed for all minor parameter directions. By performing a novel temporal SVD along policy learning path, the major and minor parameter directions are identified as the columns of right unitary matrix associated with dominant and insignificant singular values respectively. Driven by the discoveries above, we propose a simple and effective method, called Policy Path Trimming and Boosting (PPTB), as a general plug-in improvement to DRL algorithms. The key idea of PPTB is to periodically trim the policy learning path by canceling the policy updates in minor parameter directions, while boost the learning path by encouraging the advance in major directions. In experiments, we demonstrate the general and significant performance improvements brought by PPTB, when combined with TD3 and RAD in MuJoCo and DMC environments respectively.

In distributed transaction processing, atomic commit protocol (ACP) is used to ensure database consistency. With the use of commodity compute nodes and networks, failures such as system crashes and network partitioning are common. It is therefore important for ACP to dynamically adapt to the operating condition for efficiency while ensuring the consistency of the database. Existing ACPs often assume stable operating conditions, hence, they are either non-generalizable to different environments or slow in practice. In this paper, we propose a novel and practical ACP, called Failure-Aware Atomic Commit (FLAC). In essence, FLAC includes three protocols, which are specifically designed for three different environments: (i) no failure occurs, (ii) participant nodes might crash but there is no delayed connection, or (iii) both crashed nodes and delayed connection can occur. It models these environments as the failure-free, crash-failure, and network-failure robustness levels. During its operation, FLAC can monitor if any failure occurs and dynamically switch to operate the most suitable protocol, using a robustness level state machine, whose parameters are fine-tuned by reinforcement learning. Consequently, it improves both the response time and throughput, and effectively handles nodes distributed across the Internet where crash and network failures might occur. We implement FLAC in a distributed transactional key-value storage system based on Google Percolator and evaluate its performance with both a micro benchmark and a macro benchmark of real workload. The results show that FLAC achieves up to 2.22x throughput improvement and 2.82x latency speedup, compared to existing ACPs for high-contention workloads.

While imitation learning provides us with an efficient toolkit to train robots, learning skills that are robust to environment variations remains a significant challenge. Current approaches address this challenge by relying either on large amounts of demonstrations that span environment variations or on handcrafted reward functions that require state estimates. Both directions are not scalable to fast imitation. In this work, we present Fast Imitation of Skills from Humans (FISH), a new imitation learning approach that can learn robust visual skills with less than a minute of human demonstrations. Given a weak base-policy trained by offline imitation of demonstrations, FISH computes rewards that correspond to the "match" between the robot's behavior and the demonstrations. These rewards are then used to adaptively update a residual policy that adds on to the base-policy. Across all tasks, FISH requires at most twenty minutes of interactive learning to imitate demonstrations on object configurations that were not seen in the demonstrations. Importantly, FISH is constructed to be versatile, which allows it to be used across robot morphologies (e.g. xArm, Allegro, Stretch) and camera configurations (e.g. third-person, eye-in-hand). Our experimental evaluations on 9 different tasks show that FISH achieves an average success rate of 93%, which is around 3.8x higher than prior state-of-the-art methods.

In this paper, we study the Multi-Start Team Orienteering Problem (MSTOP), a mission re-planning problem where vehicles are initially located away from the depot and have different amounts of fuel. We consider/assume the goal of multiple vehicles is to travel to maximize the sum of collected profits under resource (e.g., time, fuel) consumption constraints. Such re-planning problems occur in a wide range of intelligent UAS applications where changes in the mission environment force the operation of multiple vehicles to change from the original plan. To solve this problem with deep reinforcement learning (RL), we develop a policy network with self-attention on each partial tour and encoder-decoder attention between the partial tour and the remaining nodes. We propose a modified REINFORCE algorithm where the greedy rollout baseline is replaced by a local mini-batch baseline based on multiple, possibly non-duplicate sample rollouts. By drawing multiple samples per training instance, we can learn faster and obtain a stable policy gradient estimator with significantly fewer instances. The proposed training algorithm outperforms the conventional greedy rollout baseline, even when combined with the maximum entropy objective.

To reduce the repetitive and complex work of instructors, exam paper generation (EPG) technique has become a salient topic in the intelligent education field, which targets at generating high-quality exam paper automatically according to instructor-specified assessment criteria. The current advances utilize the ability of heuristic algorithms to optimize several well-known objective constraints, such as difficulty degree, number of questions, etc., for producing optimal solutions. However, in real scenarios, considering other equally relevant objectives (e.g., distribution of exam scores, skill coverage) is extremely important. Besides, how to develop an automatic multi-objective solution that finds an optimal subset of questions from a huge search space of large-sized question datasets and thus composes a high-quality exam paper is urgent but non-trivial. To this end, we skillfully design a reinforcement learning guided Multi-Objective Exam Paper Generation framework, termed MOEPG, to simultaneously optimize three exam domain-specific objectives including difficulty degree, distribution of exam scores, and skill coverage. Specifically, to accurately measure the skill proficiency of the examinee group, we first employ deep knowledge tracing to model the interaction information between examinees and response logs. We then design the flexible Exam Q-Network, a function approximator, which automatically selects the appropriate question to update the exam paper composition process. Later, MOEPG divides the decision space into multiple subspaces to better guide the updated direction of the exam paper. Through extensive experiments on two real-world datasets, we demonstrate that MOEPG is feasible in addressing the multiple dilemmas of exam paper generation scenario.

Collision-free, goal-directed navigation in environments containing unknown static and dynamic obstacles is still a great challenge, especially when manual tuning of navigation policies or costly motion prediction needs to be avoided. In this paper, we therefore propose a subgoal-driven hierarchical navigation architecture that is trained with deep reinforcement learning and decouples obstacle avoidance and motor control. In particular, we separate the navigation task into the prediction of the next subgoal position for avoiding collisions while moving toward the final target position, and the prediction of the robot's velocity controls. By relying on 2D lidar, our method learns to avoid obstacles while still achieving goal-directed behavior as well as to generate low-level velocity control commands to reach the subgoals. In our architecture, we apply the attention mechanism on the robot's 2D lidar readings and compute the importance of lidar scan segments for avoiding collisions. As we show in simulated and real-world experiments with a Turtlebot robot, our proposed method leads to smooth and safe trajectories among humans and significantly outperforms a state-of-the-art approach in terms of success rate. A supplemental video describing our approach is available online.

In imitation and reinforcement learning, the cost of human supervision limits the amount of data that robots can be trained on. An aspirational goal is to construct self-improving robots: robots that can learn and improve on their own, from autonomous interaction with minimal human supervision or oversight. Such robots could collect and train on much larger datasets, and thus learn more robust and performant policies. While reinforcement learning offers a framework for such autonomous learning via trial-and-error, practical realizations end up requiring extensive human supervision for reward function design and repeated resetting of the environment between episodes of interactions. In this work, we propose MEDAL++, a novel design for self-improving robotic systems: given a small set of expert demonstrations at the start, the robot autonomously practices the task by learning to both do and undo the task, simultaneously inferring the reward function from the demonstrations. The policy and reward function are learned end-to-end from high-dimensional visual inputs, bypassing the need for explicit state estimation or task-specific pre-training for visual encoders used in prior work. We first evaluate our proposed algorithm on a simulated non-episodic benchmark EARL, finding that MEDAL++ is both more data efficient and gets up to 30% better final performance compared to state-of-the-art vision-based methods. Our real-robot experiments show that MEDAL++ can be applied to manipulation problems in larger environments than those considered in prior work, and autonomous self-improvement can improve the success rate by 30-70% over behavior cloning on just the expert data. Code, training and evaluation videos along with a brief overview is available at: https://architsharma97.github.io/self-improving-robots/