As humans and animals learn in the natural world, they encounter distributions of entities, situations and events that are far from uniform. Typically, a relatively small set of experiences are encountered frequently, while many important experiences occur only rarely. The highly-skewed, heavy-tailed nature of reality poses particular learning challenges that humans and animals have met by evolving specialised memory systems. By contrast, most popular RL environments and benchmarks involve approximately uniform variation of properties, objects, situations or tasks. How will RL algorithms perform in worlds (like ours) where the distribution of environment features is far less uniform? To explore this question, we develop three complementary RL environments where the agent's experience varies according to a Zipfian (discrete power law) distribution. On these benchmarks, we find that standard Deep RL architectures and algorithms acquire useful knowledge of common situations and tasks, but fail to adequately learn about rarer ones. To understand this failure better, we explore how different aspects of current approaches may be adjusted to help improve performance on rare events, and show that the RL objective function, the agent's memory system and self-supervised learning objectives can all influence an agent's ability to learn from uncommon experiences. Together, these results show that learning robustly from skewed experience is a critical challenge for applying Deep RL methods beyond simulations or laboratories, and our Zipfian environments provide a basis for measuring future progress towards this goal.
Explanations play a considerable role in human learning, especially in areas that remain major challenges for AI -- forming abstractions, and learning about the relational and causal structure of the world. Here, we explore whether reinforcement learning agents might likewise benefit from explanations. We outline a family of relational tasks that involve selecting an object that is the odd one out in a set (i.e., unique along one of many possible feature dimensions). Odd-one-out tasks require agents to reason over multi-dimensional relationships among a set of objects. We show that agents do not learn these tasks well from reward alone, but achieve >90% performance when they are also trained to generate language explaining object properties or why a choice is correct or incorrect. In further experiments, we show how predicting explanations enables agents to generalize appropriately from ambiguous, causally-confounded training, and even to meta-learn to perform experimental interventions to identify causal structure. We show that explanations help overcome the tendency of agents to fixate on simple features, and explore which aspects of explanations make them most beneficial. Our results suggest that learning from explanations is a powerful principle that could offer a promising path towards training more robust and general machine learning systems.
In naturalistic learning problems, a model's input contains a wide range of features, some useful for the task at hand, and others not. Of the useful features, which ones does the model use? Of the task-irrelevant features, which ones does the model represent? Answers to these questions are important for understanding the basis of models' decisions, for example to ensure they are equitable and unbiased, as well as for building new models that learn versatile, adaptable representations useful beyond their original training task. We study these questions using synthetic datasets in which the task-relevance of different input features can be controlled directly. We find that when two features redundantly predict the label, the model preferentially represents one, and its preference reflects what was most linearly decodable from the untrained model. Over training, task-relevant features are enhanced, and task-irrelevant features are partially suppressed. Interestingly, in some cases, an easier, weakly predictive feature can suppress a more strongly predictive, but harder one. Additionally, models trained to recognize both easy and hard features learn representations most similar to models that use only the easy feature. Further, easy features lead to more consistent representations across model runs than do hard features. Finally, models have more in common with an untrained model than with models trained on a different task. Our results highlight the complex processes that determine which features a model represents.
An important aspect of intelligence is the ability to adapt to a novel task without any direct experience (zero-shot), based on its relationship to previous tasks. Humans can exhibit this cognitive flexibility. By contrast, deep-learning models that achieve superhuman performance in specific tasks generally fail to adapt to even slight task alterations. To address this, we propose a general computational framework for adapting to novel tasks based on their relationship to prior tasks. We begin by learning vector representations of tasks. To adapt to new tasks, we propose meta-mappings, higher-order tasks that transform basic task representations. We demonstrate this framework across a wide variety of tasks and computational paradigms, ranging from regression to image classification and reinforcement learning. We compare to both human adaptability, and language-based approaches to zero-shot learning. Across these domains, meta-mapping is successful, often achieving 80-90% performance, without any data, on a novel task that directly contradicts its prior experience. We further show that using meta-mapping as a starting point can dramatically accelerate later learning on a new task, and reduce learning time and cumulative error substantially. Our results provide insight into a possible computational basis of intelligent adaptability, and offer a possible framework for modeling cognitive flexibility and building more flexible artificial intelligence.
Reinforcement learning algorithms use correlations between policies and rewards to improve agent performance. But in dynamic or sparsely rewarding environments these correlations are often too small, or rewarding events are too infrequent to make learning feasible. Human education instead relies on curricula--the breakdown of tasks into simpler, static challenges with dense rewards--to build up to complex behaviors. While curricula are also useful for artificial agents, hand-crafting them is time consuming. This has lead researchers to explore automatic curriculum generation. Here we explore automatic curriculum generation in rich, dynamic environments. Using a setter-solver paradigm we show the importance of considering goal validity, goal feasibility, and goal coverage to construct useful curricula. We demonstrate the success of our approach in rich but sparsely rewarding 2D and 3D environments, where an agent is tasked to achieve a single goal selected from a set of possible goals that varies between episodes, and identify challenges for future work. Finally, we demonstrate the value of a novel technique that guides agents towards a desired goal distribution. Altogether, these results represent a substantial step towards applying automatic task curricula to learn complex, otherwise unlearnable goals, and to our knowledge are the first to demonstrate automated curriculum generation for goal-conditioned agents in environments where the possible goals vary between episodes.
How can deep learning systems flexibly reuse their knowledge? Toward this goal, we propose a new class of challenges, and a class of architectures that can solve them. The challenges are meta-mappings, which involve systematically transforming task behaviors to adapt to new tasks zero-shot. We suggest that the key to achieving these challenges is representing the task being performed along with the computations used to perform it. We therefore draw inspiration from meta-learning and functional programming to propose a class of Embedded Meta-Learning (EML) architectures that represent both data and tasks in a shared latent space. EML architectures are applicable to any type of machine learning task, including supervised learning and reinforcement learning. We demonstrate the flexibility of these architectures by showing that they can perform meta-mappings, i.e. that they can exhibit zero-shot remapping of behavior to adapt to new tasks.
Much attention has been devoted recently to the generalization puzzle in deep learning: large, deep networks can generalize well, but existing theories bounding generalization error are exceedingly loose, and thus cannot explain this striking performance. Furthermore, a major hope is that knowledge may transfer across tasks, so that multi-task learning can improve generalization on individual tasks. However we lack analytic theories that can quantitatively predict how the degree of knowledge transfer depends on the relationship between the tasks. We develop an analytic theory of the nonlinear dynamics of generalization in deep linear networks, both within and across tasks. In particular, our theory provides analytic solutions to the training and testing error of deep networks as a function of training time, number of examples, network size and initialization, and the task structure and SNR. Our theory reveals that deep networks progressively learn the most important task structure first, so that generalization error at the early stopping time primarily depends on task structure and is independent of network size. This suggests any tight bound on generalization error must take into account task structure, and explains observations about real data being learned faster than random data. Intriguingly our theory also reveals the existence of a learning algorithm that proveably out-performs neural network training through gradient descent. Finally, for transfer learning, our theory reveals that knowledge transfer depends sensitively, but computably, on the SNRs and input feature alignments of pairs of tasks.
Standard deep learning systems require thousands or millions of examples to learn a concept, and cannot integrate new concepts easily. By contrast, humans have an incredible ability to do one-shot or few-shot learning. For instance, from just hearing a word used in a sentence, humans can infer a great deal about it, by leveraging what the syntax and semantics of the surrounding words tells us. Here, we draw inspiration from this to highlight a simple technique by which deep recurrent networks can similarly exploit their prior knowledge to learn a useful representation for a new word from little data. This could make natural language processing systems much more flexible, by allowing them to learn continually from the new words they encounter.