The reviewers consider the task suite, the memory-augmented model, and the evaluations to be solid contributions. Please be sure to release the task suite...

Episodic control enables sample efficiency in reinforcement learning by recalling past experiences from an episodic memory. We propose a new model-based episodic memory of trajectories addressing current limitations of episodic control. Built upon the memory, we construct a complementary learning model via a dynamic hybrid control unifying model-based, episodic and habitual learning into a single architecture. Experiments demonstrate that our model allows significantly faster and better learning than other strong reinforcement learning agents across a variety of environments including stochastic and non-Markovian settings.

Social learning, a cornerstone of cultural evolution, enables individuals to acquire knowledge by observing and imitating others. At the heart of its efficacy lies episodic memory, which encodes specific behavioral sequences to facilitate learning and decision-making. This study explores the interrelation between episodic memory and social learning in collective foraging. Using Sequential Episodic Control (SEC) agents capable of sharing complete behavioral sequences stored in episodic memory, we investigate how variations in the frequency and fidelity of social learning influence collaborative foraging performance. Furthermore, we analyze the effects of social learning on the content and distribution of episodic memories across the group. High-fidelity social learning is shown to consistently enhance resource collection efficiency and distribution, with benefits sustained across memory lengths. In contrast, low-fidelity learning fails to outperform nonsocial learning, spreading diverse but ineffective mnemonic patterns. Novel analyses using mnemonic metrics reveal that high-fidelity social learning also fosters mnemonic group alignment and equitable resource distribution, while low-fidelity conditions increase mnemonic diversity without translating to performance gains. Additionally, we identify an optimal range for episodic memory length in this task, beyond which performance plateaus. These findings underscore the critical effects of social learning on mnemonic group alignment and distribution and highlight the potential of neurocomputational models to probe the cognitive mechanisms driving cultural evolution.

Humans navigate unfamiliar environments using episodic simulation and episodic memory, which facilitate a deeper understanding of the complex relationships between environments and objects. Developing an imaginative memory system inspired by human mechanisms can enhance the navigation performance of embodied agents in unseen environments. However, existing Vision-and-Language Navigation (VLN) agents lack a memory mechanism of this kind. To address this, we propose a novel architecture that equips agents with a reality-imagination hybrid memory system. This system enables agents to maintain and expand their memory through both imaginative mechanisms and navigation actions. Additionally, we design tailored pre-training tasks to develop the agent's imaginative capabilities. Our agent can imagine high-fidelity RGB images for future scenes, achieving state-of-the-art result in Success rate weighted by Path Length (SPL).

In this paper, we study the problem of generating structured objects that conform to a complex schema, with intricate dependencies between the different components (facets) of the object. The facets of the object (attributes, fields, columns, properties) can be a mix of short, structured, type-constrained facts, or long natural-language descriptions. The object has to be self-consistent between the different facets in the redundant information it carries (relative consistency), while being grounded with respect to world knowledge (absolute consistency). We frame the problem as a Language Modeling problem (Structured Object Language Modeling) and train an LLM to perform the task natively, without requiring instructions or prompt-engineering. We propose a self-supervised denoising method to train the model from an existing dataset of such objects. The input query can be the existing object itself, in which case the model acts as a regenerator, completing, correcting, normalizing the input, or any unstructured blurb to be structured. We show that the self-supervised denoising training provides a strong baseline, and that additional supervised fine-tuning with small amount of human demonstrations leads to further improvement. Experimental results show that the proposed method matches or outperforms prompt-engineered general-purpose state-of-the-art LLMs (Claude 3, Mixtral-8x7B), while being order-of-magnitude more cost-efficient.

This review examines theoretical assumptions and computational models of event comprehension, tracing the evolution from discourse comprehension theories to contemporary event cognition frameworks. The review covers key discourse comprehension accounts, including Construction-Integration, Event Indexing, Causal Network, and Resonance models, highlighting their contributions to understanding cognitive processes in comprehension. I then discuss contemporary theoretical frameworks of event comprehension, including Event Segmentation Theory (Zacks et al., 2007), the Event Horizon Model (Radvansky & Zacks, 2014), and Hierarchical Generative Framework (Kuperberg, 2021), which emphasize prediction, causality, and multilevel representations in event understanding. Building on these theories, I evaluate five computational models of event comprehension: REPRISE (Butz et al., 2019), Structured Event Memory (SEM; Franklin et al., 2020), the Lu model (Lu et al., 2022), the Gumbsch model (Gumbsch et al., 2022), and the Elman and McRae model (2019). The analysis focuses on their approaches to hierarchical processing, prediction mechanisms, and representation learning. Key themes that emerge include the use of hierarchical structures as inductive biases, the importance of prediction in comprehension, and diverse strategies for learning event dynamics. The review identifies critical areas for future research, including the need for more sophisticated approaches to learning structured representations, integrating episodic memory mechanisms, and developing adaptive updating algorithms for working event models. By synthesizing insights from both theoretical frameworks and computational implementations, this review aims to advance our understanding of human event comprehension and guide future modeling efforts in cognitive science.

We introduce a lifelong language learning setup where a model needs to learn from a stream of text examples without any dataset identifier. We propose an episodic memory model that performs sparse experience replay and local adaptation to mitigate catastrophic forgetting in this setup. Experiments on text classification and question answering demonstrate the complementary benefits of sparse experience replay and local adaptation to allow the model to continuously learn from new datasets. We also show that the space complexity of the episodic memory module can be reduced significantly ( 50-90%) by randomly choosing which examples to store in memory with a minimal decrease in performance. We consider an episodic memory component as a crucial building block of general linguistic intelligence and see our model as a first step in that direction.

Generative Adversarial Networks (GANs) struggle to generate structured objects like molecules and game maps. The issue is that structured objects must satisfy hard requirements (e.g., molecules must be chemically valid) that are difficult to acquire from examples alone. As a remedy, we propose Constrained Adversarial Networks (CANs), an extension of GANs in which the constraints are embedded into the model during training. This is achieved by penalizing the generator proportionally to the mass it allocates to invalid structures. In contrast to other generative models, CANs support efficient inference of valid structures (with high probability) and allows to turn on and off the learned constraints at inference time.

Current deep neural networks can achieve remarkable performance on a single task. However, when the deep neural network is continually trained on a sequence of tasks, it seems to gradually forget the previous learned knowledge. This phenomenon is referred to as catastrophic forgetting and motivates the field called lifelong learning. Recently, episodic memory based approaches such as GEM and A-GEM have shown remarkable performance. In this paper, we provide the first unified view of episodic memory based approaches from an optimization's perspective.

Memory is an important aspect of intelligence and plays a role in many deep reinforcement learning models. However, little progress has been made in understanding when specific memory systems help more than others and how well they generalize. The field also has yet to see a prevalent consistent and rigorous approach for evaluating agent performance on holdout data. In this paper, we aim to develop a comprehensive methodology to test different kinds of memory in an agent and assess how well the agent can apply what it learns in training to a holdout set that differs from the training set along dimensions that we suggest are relevant for evaluating memory-specific generalization. To that end, we first construct a diverse set of memory tasks that allow us to evaluate test-time generalization across multiple dimensions. Second, we develop and perform multiple ablations on an agent architecture that combines multiple memory systems, observe its baseline models, and investigate its performance against the task suite.