Generating explanations for reinforcement learning (RL) is challenging as actions may produce long-term effects on the future. In this paper, we develop a novel framework for explainable RL by learning a causal world model without prior knowledge of the causal structure of the environment. The model captures the influence of actions, allowing us to interpret the long-term effects of actions through causal chains, which present how actions influence environmental variables and finally lead to rewards. Different from most explanatory models which suffer from low accuracy, our model remains accurate while improving explainability, making it applicable in model-based learning. As a result, we demonstrate that our causal model can serve as the bridge between explainability and learning.

Existing combinatorial search methods are often complex and require some level of expertise. This work introduces a simple and efficient deep learning method for solving combinatorial problems with a predefined goal, represented by Rubik's Cube. We demonstrate that, for such problems, training a deep neural network on random scrambles branching from the goal state is sufficient to achieve near-optimal solutions. When tested on Rubik's Cube, 15 Puzzle, and 7$\times$7 Lights Out, our method outperformed the previous state-of-the-art method DeepCubeA, improving the trade-off between solution optimality and computational cost, despite significantly less training data. Furthermore, we investigate the scaling law of our Rubik's Cube solver with respect to model size and training data volume.

Fact-checking real-world claims often requires collecting multiple pieces of evidence and applying complex multi-step reasoning. In this paper, we present Program-Guided Fact-Checking (ProgramFC), a novel fact-checking model that decomposes complex claims into simpler sub-tasks that can be solved using a shared library of specialized functions. We first leverage the in-context learning ability of large language models to generate reasoning programs to guide the verification process. Afterward, we execute the program by delegating each sub-task to the corresponding sub-task handler. This process makes our model both explanatory and data-efficient, providing clear explanations of its reasoning process and requiring minimal training data. We evaluate ProgramFC on two challenging fact-checking datasets and show that it outperforms seven fact-checking baselines across different settings of evidence availability, with explicit output programs that benefit human debugging. Our codes and data are publicly available at https://github.com/mbzuai-nlp/ProgramFC.

Much attention and concern has been raised recently about bias and the use of machine learning algorithms in healthcare, especially as it relates to perpetuating racial discrimination and health disparities. Following an initial system dynamics workshop at the Data for Black Lives II conference hosted at MIT in January of 2019, a group of conference participants interested in building capabilities to use system dynamics to understand complex societal issues convened monthly to explore issues related to racial bias in AI and implications for health disparities through qualitative and simulation modeling. In this paper we present results and insights from the modeling process and highlight the importance of centering the discussion of data and healthcare on people and their experiences with healthcare and science, and recognizing the societal context where the algorithm is operating. Collective memory of community trauma, through deaths attributed to poor healthcare, and negative experiences with healthcare are endogenous drivers of seeking treatment and experiencing effective care, which impact the availability and quality of data for algorithms. These drivers have drastically disparate initial conditions for different racial groups and point to limited impact of focusing solely on improving diagnostic algorithms for achieving better health outcomes for some groups.

Interactive Natural Language Processing (iNLP) has emerged as a novel paradigm within the field of NLP, aimed at addressing limitations in existing frameworks while aligning with the ultimate goals of artificial intelligence. This paradigm considers language models as agents capable of observing, acting, and receiving feedback iteratively from external entities. Specifically, language models in this context can: (1) interact with humans for better understanding and addressing user needs, personalizing responses, aligning with human values, and improving the overall user experience; (2) interact with knowledge bases for enriching language representations with factual knowledge, enhancing the contextual relevance of responses, and dynamically leveraging external information to generate more accurate and informed responses; (3) interact with models and tools for effectively decomposing and addressing complex tasks, leveraging specialized expertise for specific subtasks, and fostering the simulation of social behaviors; and (4) interact with environments for learning grounded representations of language, and effectively tackling embodied tasks such as reasoning, planning, and decision-making in response to environmental observations. This paper offers a comprehensive survey of iNLP, starting by proposing a unified definition and framework of the concept. We then provide a systematic classification of iNLP, dissecting its various components, including interactive objects, interaction interfaces, and interaction methods. We proceed to delve into the evaluation methodologies used in the field, explore its diverse applications, scrutinize its ethical and safety issues, and discuss prospective research directions. This survey serves as an entry point for researchers who are interested in this rapidly evolving area and offers a broad view of the current landscape and future trajectory of iNLP.

Overlapping instruction subsets derived from human originated code have previously been shown to dramatically shrink the inductive programming search space, often by many orders of magnitude. Here we extend the instruction subset approach to consider direct instruction-instruction applications (or instruction digrams) as an additional search heuristic for inductive programming. In this study we analyse the frequency distribution of instruction digrams in a large sample of open source code. This indicates that the instruction digram distribution is highly skewed with over 93% of possible instruction digrams not represnted in the code sample. We demonstrate that instruction digrams can be used to constrain instruction selection during search, further reducing size of the the search space, in some cases by several orders of magnitude. This significantly increases the size of programs that can be generated using search based inductive programming techniques. We discuss the results and provide some suggestions for further work.

Neural language models (LMs) represent facts about the world described by text. Sometimes these facts derive from training data (in most LMs, a representation of the word "banana" encodes the fact that bananas are fruits). Sometimes facts derive from input text itself (a representation of the sentence "I poured out the bottle" encodes the fact that the bottle became empty). We describe REMEDI, a method for learning to map statements in natural language to fact encodings in an LM's internal representation system. REMEDI encodings can be used as knowledge editors: when added to LM hidden representations, they modify downstream generation to be consistent with new facts. REMEDI encodings may also be used as probes: when compared to LM representations, they reveal which properties LMs already attribute to mentioned entities, in some cases making it possible to predict when LMs will generate outputs that conflict with background knowledge or input text. REMEDI thus links work on probing, prompting, and LM editing, and offers steps toward general tools for fine-grained inspection and control of knowledge in LMs.

Physics is a field of science that has traditionally used the scientific method to answer questions about why natural phenomena occur and to make testable models that explain the phenomena. Discovering equations, laws and principles that are invariant, robust and causal explanations of the world has been fundamental in physical sciences throughout the centuries. Discoveries emerge from observing the world and, when possible, performing interventional studies in the system under study. With the advent of big data and the use of data-driven methods, causal and equation discovery fields have grown and made progress in computer science, physics, statistics, philosophy, and many applied fields. All these domains are intertwined and can be used to discover causal relations, physical laws, and equations from observational data. This paper reviews the concepts, methods, and relevant works on causal and equation discovery in the broad field of Physics and outlines the most important challenges and promising future lines of research. We also provide a taxonomy for observational causal and equation discovery, point out connections, and showcase a complete set of case studies in Earth and climate sciences, fluid dynamics and mechanics, and the neurosciences. This review demonstrates that discovering fundamental laws and causal relations by observing natural phenomena is being revolutionised with the efficient exploitation of observational data, modern machine learning algorithms and the interaction with domain knowledge. Exciting times are ahead with many challenges and opportunities to improve our understanding of complex systems.

We present Locksynth, a tool that automatically derives synchronization needed for destructive updates to concurrent data structures that involve a constant number of shared heap memory write operations. Locksynth serves as the implementation of our prior work on deriving abstract synchronization code. Designing concurrent data structures involves inferring correct synchronization code starting with a prior understanding of the sequential data structure's operations. Further, an understanding of shared memory model and the synchronization primitives is also required. The reasoning involved transforming a sequential data structure into its concurrent version can be performed using Answer Set Programming and we mechanized our approach in previous work. The reasoning involves deduction and abduction that can be succinctly modeled in ASP. We assume that the abstract sequential code of the data structure's operations is provided, alongside axioms that describe concurrent behavior. This information is used to automatically derive concurrent code for that data structure, such as dictionary operations for linked lists and binary search trees that involve a constant number of destructive update operations. We also are able to infer the correct set of locks (but not code synthesis) for external height-balanced binary search trees that involve left/right tree rotations. Locksynth performs the analyses required to infer correct sets of locks and as a final step, also derives the C++ synchronization code for the synthesized data structures. We also provide a performance analysis of the C++ code synthesized by Locksynth with the hand-crafted versions available from the Synchrobench microbenchmark suite. To the best of our knowledge, our tool is the first to employ ASP as a backend reasoner to perform concurrent data structure synthesis.

Trustworthiness of artificially intelligent agents is vital for the acceptance of human-machine teaming in industrial manufacturing environments. Predictable behaviours and explainable (and understandable) rationale allow humans collaborating with (and building) these agents to understand their motivations and therefore validate decisions that are made. To that aim, we make use of G\"ardenfors's cognitively inspired Conceptual Space framework to represent the agent's knowledge using concepts as convex regions in a space spanned by inherently comprehensible quality dimensions. A simple typicality quantification model is built on top of it to determine fuzzy category membership and classify instances interpretably. We apply it on a use case from the manufacturing domain, using objects' physical properties obtained from cobots' onboard sensors and utilisation properties from crowdsourced commonsense knowledge available at public knowledge bases. Such flexible knowledge representation based on property decomposition allows for data-efficient representation learning of typically highly specialist or specific manufacturing artefacts. In such a setting, traditional data-driven (e.g., computer vision-based) classification approaches would struggle due to training data scarcity. This allows for comprehensibility of an AI agent's acquired knowledge by the human collaborator thus contributing to trustworthiness. We situate our approach within an existing explainability framework specifying explanation desiderata. We provide arguments for our system's applicability and appropriateness for different roles of human agents collaborating with the AI system throughout its design, validation, and operation.