Open-Set Domain Adaptation (OSDA) assumes that a target domain contains unknown classes, which are not discovered in a source domain. Existing domain adversarial learning methods are not suitable for OSDA because distribution matching with $\textit{unknown}$ classes leads to negative transfer. Previous OSDA methods have focused on matching the source and the target distribution by only utilizing $\textit{known}$ classes. However, this $\textit{known}$-only matching may fail to learn the target-$\textit{unknown}$ feature space. Therefore, we propose Unknown-Aware Domain Adversarial Learning (UADAL), which $\textit{aligns}$ the source and the target-$\textit{known}$ distribution while simultaneously $\textit{segregating}$ the target-$\textit{unknown}$ distribution in the feature alignment procedure. We provide theoretical analyses on the optimized state of the proposed $\textit{unknown-aware}$ feature alignment, so we can guarantee both $\textit{alignment}$ and $\textit{segregation}$ theoretically. Empirically, we evaluate UADAL on the benchmark datasets, which shows that UADAL outperforms other methods with better feature alignments by reporting state-of-the-art performances.

In this paper, we establish the global optimality and convergence rate of an off-policy actor critic algorithm in the tabular setting without using density ratio to correct the discrepancy between the state distribution of the behavior policy and that of the target policy. Our work goes beyond existing works on the optimality of policy gradient methods in that existing works use the exact policy gradient for updating the policy parameters while we use an approximate and stochastic update step. Our update step is not a gradient update because we do not use a density ratio to correct the state distribution, which aligns well with what practitioners do. Our update is approximate because we use a learned critic instead of the true value function. Our update is stochastic because at each step the update is done for only the current state action pair. Moreover, we remove several restrictive assumptions from existing works in our analysis. Central to our work is the finite sample analysis of a generic stochastic approximation algorithm with time-inhomogeneous update operators on time-inhomogeneous Markov chains, based on its uniform contraction properties.

Adaptive curricula in reinforcement learning (RL) have proven effective for producing policies robust to discrepancies between the train and test environment. Recently, the Unsupervised Environment Design (UED) framework generalized RL curricula to generating sequences of entire environments, leading to new methods with robust minimax regret properties. Problematically, in partially-observable or stochastic settings, optimal policies may depend on the ground-truth distribution over aleatoric parameters of the environment in the intended deployment setting, while curriculum learning necessarily shifts the training distribution. We formalize this phenomenon as curriculum-induced covariate shift (CICS), and describe how its occurrence in aleatoric parameters can lead to suboptimal policies. Directly sampling these parameters from the ground-truth distribution avoids the issue, but thwarts curriculum learning. We propose SAMPLR, a minimax regret UED method that optimizes the ground-truth utility function, even when the underlying training data is biased due to CICS. We prove, and validate on challenging domains, that our approach preserves optimality under the ground-truth distribution, while promoting robustness across the full range of environment settings.

I show that when the money supply accelerates, the learning agents only adjust their actions, which include consumption and demand for real balance, after gathering learning experience for many periods. This delayed adjustments leads to low returns during transition periods. Once they start adjusting to the new environment, their welfare improves. Their changes in beliefs and actions lead to temporary inflation volatility. I also show that, 1. the AI agents who explores their environment more adapt to the policy regime change quicker, which leads to welfare improvements and less inflation volatility, and 2. the AI agents who have experienced a structural change adjust their beliefs and behaviours quicker than an inexperienced learning agent.

Climate Change is an incredibly complicated problem that humanity faces. When many variables interact with each other, it can be difficult for humans to grasp the causes and effects of the very large-scale problem of climate change. The climate is a dynamical system, where small changes can have considerable and unpredictable repercussions in the long term. Understanding how to nudge this system in the right ways could help us find creative solutions to climate change. In this research, we combine Deep Reinforcement Learning and a World-Earth system model to find, and explain, creative strategies to a sustainable future. This is an extension of the work from Strnad et al. where we extend on the method and analysis, by taking multiple directions. We use four different Reinforcement Learning agents varying in complexity to probe the environment in different ways and to find various strategies. The environment is a low-complexity World Earth system model where the goal is to reach a future where all the energy for the economy is produced by renewables by enacting different policies. We use a reward function based on planetary boundaries that we modify to force the agents to find a wider range of strategies. To favour applicability, we slightly modify the environment, by injecting noise and making it fully observable, to understand the impacts of these factors on the learning of the agents.

Building generally capable agents is a grand challenge for deep reinforcement learning (RL). To approach this challenge practically, we outline two key desiderata: 1) to facilitate generalization, exploration should be task agnostic; 2) to facilitate scalability, exploration policies should collect large quantities of data without costly centralized retraining. Combining these two properties, we introduce the reward-free deployment efficiency setting, a new paradigm for RL research. We then present CASCADE, a novel approach for self-supervised exploration in this new setting. CASCADE seeks to learn a world model by collecting data with a population of agents, using an information theoretic objective inspired by Bayesian Active Learning. CASCADE achieves this by specifically maximizing the diversity of trajectories sampled by the population through a novel cascading objective. We provide theoretical intuition for CASCADE which we show in a tabular setting improves upon na\"ive approaches that do not account for population diversity. We then demonstrate that CASCADE collects diverse task-agnostic datasets and learns agents that generalize zero-shot to novel, unseen downstream tasks on Atari, MiniGrid, Crafter and the DM Control Suite. Code and videos are available at https://ycxuyingchen.github.io/cascade/

Recent advances in TCP congestion control (CC) have achieved tremendous success with deep reinforcement learning (RL) approaches, which use feedforward neural networks (NN) to learn complex environment conditions and make better decisions. However, such "black-box" policies lack interpretability and reliability, and often, they need to operate outside the traditional TCP datapath due to the use of complex NNs. This paper proposes a novel two-stage solution to achieve the best of both worlds: first to train a deep RL agent, then distill its (over-)parameterized NN policy into white-box, light-weight rules in the form of symbolic expressions that are much easier to understand and to implement in constrained environments. At the core of our proposal is a novel symbolic branching algorithm that enables the rule to be aware of the context in terms of various network conditions, eventually converting the NN policy into a symbolic tree. The distilled symbolic rules preserve and often improve performance over state-of-the-art NN policies while being faster and simpler than a standard neural network. We validate the performance of our distilled symbolic rules on both simulation and emulation environments.

Actor learning and critic learning are two components of the outstanding and mostly used Deep Deterministic Policy Gradient (DDPG) reinforcement learning method. Since actor and critic learning plays a significant role in the overall robot's learning, the performance of the DDPG approach is relatively sensitive and unstable as a result. We propose a multi-actor-critic DDPG for reliable actor-critic learning to further enhance the performance and stability of DDPG. This multi-actor-critic DDPG is then integrated with Hindsight Experience Replay (HER) to form our new deep learning framework called AACHER. AACHER uses the average value of multiple actors or critics to substitute the single actor or critic in DDPG to increase resistance in the case when one actor or critic performs poorly. Numerous independent actors and critics can also gain knowledge from the environment more broadly. We implemented our proposed AACHER on goal-based environments: AuboReach, FetchReach-v1, FetchPush-v1, FetchSlide-v1, and FetchPickAndPlace-v1. For our experiments, we used various instances of actor/critic combinations, among which A10C10 and A20C20 were the best-performing combinations. Overall results show that AACHER outperforms the traditional algorithm (DDPG+HER) in all of the actor/critic number combinations that are used for evaluation. When used on FetchPickAndPlace-v1, the performance boost for A20C20 is as high as roughly 3.8 times the success rate in DDPG+HER.

Intelligence agents and multi-agent systems play important roles in scenes like the control system of grouped drones, and multi-agent navigation and obstacle avoidance which is the foundational function of advanced application has great importance. In multi-agent navigation and obstacle avoidance tasks, the decision-making interactions and dynamic changes of agents are difficult for traditional route planning algorithms or reinforcement learning algorithms with the increased complexity of the environment. The classical multi-agent reinforcement learning algorithm, Multi-agent deep deterministic policy gradient(MADDPG), solved precedent algorithms' problems of having unstationary training process and unable to deal with environment randomness. However, MADDPG ignored the temporal message hidden beneath agents' interaction with the environment. Besides, due to its CTDE technique which let each agent's critic network to calculate over all agents' action and the whole environment information, it lacks ability to scale to larger amount of agents. To deal with MADDPG's ignorance of the temporal information of the data, this article proposes a new algorithm called MADDPG-LSTMactor, which combines MADDPG with Long short term memory (LSTM). By using agent's observations of continuous timesteps as the input of its policy network, it allows the LSTM layer to process the hidden temporal message. Experimental result demonstrated that this algorithm had better performance in scenarios where the amount of agents is small. Besides, to solve MADDPG's drawback of not being efficient in scenarios where agents are too many, this article puts forward a light-weight MADDPG (MADDPG-L) algorithm, which simplifies the input of critic network. The result of experiments showed that this algorithm had better performance than MADDPG when the amount of agents was large.

Predictive coding has emerged as a prominent model of how the brain learns through predictions, anticipating the importance accorded to predictive learning in recent AI architectures such as transformers. Here we propose a new framework for predictive coding called active predictive coding which can learn hierarchical world models and solve two radically different open problems in AI: (1) how do we learn compositional representations, e.g., part-whole hierarchies, for equivariant vision? and (2) how do we solve large-scale planning problems, which are hard for traditional reinforcement learning, by composing complex action sequences from primitive policies? Our approach exploits hypernetworks, self-supervised learning and reinforcement learning to learn hierarchical world models that combine task-invariant state transition networks and task-dependent policy networks at multiple abstraction levels. We demonstrate the viability of our approach on a variety of vision datasets (MNIST, FashionMNIST, Omniglot) as well as on a scalable hierarchical planning problem. Our results represent, to our knowledge, the first demonstration of a unified solution to the part-whole learning problem posed by Hinton, the nested reference frames problem posed by Hawkins, and the integrated state-action hierarchy learning problem in reinforcement learning.