Interpretability in machine learning is the ability to explain or to present in understandable terms to a human [8]. Interpretability is particularly important when, for example the goal of the user is to gain knowledge from some form of explanations about the data or process through machine learning models, or when making high-stakes decisions based on the outputs from the machine learning models where the user has to be able to trust the models. In this work we address the problem of explaining and understanding tree-ensemble learners by extracting meaningful rules from them. This problem is of practical relevance in business domains where the understanding of the behavior of high-performing machine learning models and extraction of knowledge in human readable form can aid users in the decision making process. We use Answer Set Programming (ASP) [14, 22] to generate rule sets from tree-ensembles.

Neural embedding-based machine learning models have shown promise for predicting novel links in knowledge graphs. Unfortunately, their practical utility is diminished by their lack of interpretability. Recently, the fully interpretable, rule-based algorithm AnyBURL yielded highly competitive results on many general-purpose link prediction benchmarks. However, current approaches for aggregating predictions made by multiple rules are affected by redundancies. We improve upon AnyBURL by introducing the SAFRAN rule application framework, which uses a novel aggregation approach called Non-redundant Noisy-OR that detects and clusters redundant rules prior to aggregation. SAFRAN yields new state-of-the-art results for fully interpretable link prediction on the established general-purpose benchmarks FB15K-237, WN18RR and YAGO3-10. Furthermore, it exceeds the results of multiple established embedding-based algorithms on FB15K-237 and WN18RR and narrows the gap between rule-based and embedding-based algorithms on YAGO3-10.

We are interested in understanding the underlying generation process for long sequences of symbolic events. To do so, we propose COSSU, an algorithm to mine small and meaningful sets of sequential rules. The rules are selected using an MDL-inspired criterion that favors compactness and relies on a novel rule-based encoding scheme for sequences. Our evaluation shows that COSSU can successfully retrieve relevant sets of closed sequential rules from a long sequence. Such rules constitute an interpretable model that exhibits competitive accuracy for the tasks of next-element prediction and classification.

Decisions and the underlying rules are indispensable for driving process execution during runtime, i.e., for routing process instances at alternative branches based on the values of process data. Decision rules can comprise unary data conditions, e.g., age > 40, binary data conditions where the relation between two or more variables is relevant, e.g. temperature1 < temperature2, and more complex conditions that refer to, for example, parts of a medical image. Decision discovery aims at automatically deriving decision rules from process event logs. Existing approaches focus on the discovery of unary, or in some instances binary data conditions. The discovered decision rules are usually evaluated using accuracy, but not with regards to their semantics and meaningfulness, although this is crucial for validation and the subsequent implementation/adaptation of the decision rules. Hence, this paper compares three decision mining approaches, i.e., two existing ones and one newly described approach, with respect to the meaningfulness of their results. For comparison, we use one synthetic data set for a realistic manufacturing case and the two real-world BPIC 2017/2020 logs. The discovered rules are discussed with regards to their semantics and meaningfulness.

Recent years have seen an increasing need for gender-neutral and inclusive language. Within the field of NLP, there are various mono- and bilingual use cases where gender inclusive language is appropriate, if not preferred due to ambiguity or uncertainty in terms of the gender of referents. In this work, we present a rule-based and a neural approach to gender-neutral rewriting for English along with manually curated synthetic data (WinoBias+) and natural data (OpenSubtitles and Reddit) benchmarks. A detailed manual and automatic evaluation highlights how our NeuTral Rewriter, trained on data generated by the rule-based approach, obtains word error rates (WER) below 0.18% on synthetic, in-domain and out-domain test sets.

A key component of mathematical reasoning is the ability to formulate interesting conjectures about a problem domain at hand. In this paper, we give a brief overview of a theory exploration system called QuickSpec, which is able to automatically discover interesting conjectures about a given set of functions. QuickSpec works by interleaving term generation with random testing to form candidate conjectures. This is made tractable by starting from small sizes and ensuring that only terms that are irreducible with respect to already discovered conjectures are considered. QuickSpec has been successfully applied to generate lemmas for automated inductive theorem proving as well as to generate specifications of functional programs. We give an overview of typical use-cases of QuickSpec, as well as demonstrating how to easily connect it to a theorem prover of the user's choice.

A variation of the problems posed by black-box decision-making is the experience of researchers at Mount Sinai Hospital in New York, in applying a learning system to the hospitals' database of records on some 700,000 individuals. The resulting learning system, called Deep Patient, turned out to be very good at predicting disease. It even appeared to anticipate the onset of psychiatric disorders like schizophrenia, which is difficult for physicians to predict, quite well. "Deep Patient offers no clue as to how it does this," say the authors, referencing Joel Dudley, former leader of the Mount Sinai team, now chief scientific officer at Tempus Labs, which advances precision medicine through the practical application of AI in healthcare.

When I wrote recently that California's recall election process was terribly flawed and in need of serious reform, the angry messages came flowing in, calling me a hypocrite. The writers didn't believe for a second that I objected to the recall on principle -- they assumed that as a loyal Democrat, I was just shilling for Gov. Gavin Newsom. "You're in his pocket," said one dismissive tweet. Would I still be vehemently opposed to the recall if, instead of being used against a Democratic governor, it was targeting a Trump-supporting right-wing governor -- someone who, say, was unleashing the fossil fuel industry, hoping to do away with the minimum wage and fighting mask and vaccine mandates? Would I still feel the recall was a troubling, badly structured, overused, undemocratic tool that should be reformed or abolished?

In recent years, medical resources in China have been in a state of short supply. Doctors from the top hospitals in modern cities have to face tedious consultation or surgical work every day. Meanwhile, some doctors also need to do scientific research related work for the development of medicine. However, in doctors' daily work, a large part of the work is cumbersome but easy to use IT systems to improve efficiency, such as the management of medical terms, the collection of medical knowledge, and so on. The organization of medical knowledge is mostly for daily diagnosis and treatment in accordance with certain gold standards or guidelines, and many of them are based on some established rules to make inference.

Autonomous vehicles need to model the behavior of surrounding human driven vehicles to be safe and efficient traffic participants. Existing approaches to modeling human driving behavior have relied on both data-driven and rule-based methods. While data-driven models are more expressive, rule-based models are interpretable, which is an important requirement for safety-critical domains like driving. However, rule-based models are not sufficiently representative of data, and data-driven models are yet unable to generate realistic traffic simulation due to unrealistic driving behavior such as collisions. In this paper, we propose a methodology that combines rule-based modeling with data-driven learning. While the rules are governed by interpretable parameters of the driver model, these parameters are learned online from driving demonstration data using particle filtering. We perform driver modeling experiments on the task of highway driving and merging using data from three real-world driving demonstration datasets. Our results show that driver models based on our hybrid rule-based and data-driven approach can accurately capture real-world driving behavior. Further, we assess the realism of the driving behavior generated by our model by having humans perform a driving Turing test, where they are asked to distinguish between videos of real driving and those generated using our driver models.