Spoken question answering (SQA) is a challenging task that requires the machine to fully understand the complex spoken documents. Automatic speech recognition (ASR) plays a significant role in the development of QA systems. However, the recent work shows that ASR systems generate highly noisy transcripts, which critically limit the capability of machine comprehension on the SQA task. To address the issue, we present a novel distillation framework. Specifically, we devise a training strategy to perform knowledge distillation (KD) from spoken documents and written counterparts. Our work makes a step towards distilling knowledge from the language model as a supervision signal to lead to better student accuracy by reducing the misalignment between automatic and manual transcriptions. Experiments demonstrate that our approach outperforms several state-of-the-art language models on the Spoken-SQuAD dataset.

Unsupervised and self-supervised learning approaches have become a crucial tool to learn representations for downstream prediction tasks. While these approaches are widely used in practice and achieve impressive empirical gains, their theoretical understanding largely lags behind. Towards bridging this gap, we present a unifying perspective where several such approaches can be viewed as imposing a regularization on the representation via a learnable function using unlabeled data. We propose a discriminative theoretical framework for analyzing the sample complexity of these approaches, which generalizes the framework of (Balcan and Blum, 2010) to allow learnable regularization functions. Our sample complexity bounds show that, with carefully chosen hypothesis classes to exploit the structure in the data, these learnable regularization functions can prune the hypothesis space, and help reduce the amount of labeled data needed. We then provide two concrete examples of functional regularization, one using auto-encoders and the other using masked self-supervision, and apply our framework to quantify the reduction in the sample complexity bound of labeled data. We also provide complementary empirical results to support our analysis.

It has been recently demonstrated that multi-generational self-distillation can improve generalization [11]. Despite this intriguing observation, reasons for the enhancement remain poorly understood. In this paper, we first demonstrate experimentally that the improved performance of multi-generational self-distillation is in part associated with the increasing diversity in teacher predictions. With this in mind, we offer a new interpretation for teacher-student training as amortized MAP estimation, such that teacher predictions enable instance-specific regularization. Our framework allows us to theoretically relate self-distillation to label smoothing, a commonly used technique that regularizes predictive uncertainty, and suggests the importance of predictive diversity in addition to predictive uncertainty. We present experimental results using multiple datasets and neural network architectures that, overall, demonstrate the utility of predictive diversity. Finally, we propose a novel instance-specific label smoothing technique that promotes predictive diversity without the need for a separately trained teacher model. We provide an empirical evaluation of the proposed method, which, we find, often outperforms classical label smoothing.

Process Mining has recently gained popularity in healthcare due to its potential to provide a transparent, objective and data-based view on processes. Conformance checking is a sub-discipline of process mining that has the potential to answer how the actual process executions deviate from existing guidelines. In this work, we analyze a medical training process for a surgical procedure. Ten students were trained to install a Central Venous Catheters (CVC) with ultrasound. Event log data was collected directly after instruction by the supervisors during a first test run and additionally after a subsequent individual training phase. In order to provide objective performance measures, we formulate an optimal, global sequence alignment problem inspired by approaches in bioinformatics. Therefore, we use the Petri net model representation of the medical process guideline to simulate a representative set of guideline conform sequences. Next, we calculate the optimal, global sequence alignment of the recorded and simulated event logs. Finally, the output measures and visualization of aligned sequences are provided for objective feedback.

Introductory hands-on courses such as our smartphone-based coding courses, SuaCode require a lot of support for students to accomplish learning goals. Online environments make it even more difficult to get assistance especially more recently because of COVID-19. Given the multilingual context of our students (learners across 38 African countries), in this work, we developed an AI Teaching Assistant (Kwame) that provides answers to students' coding questions from our SuaCode courses in English and French. Kwame is a Sentence-BERT(SBERT)-based question-answering (QA) system that we trained and evaluated using question-answer pairs created from our course's quizzes and students' questions in past cohorts. It finds the paragraph most semantically similar to the question via cosine similarity. We compared the system with TF-IDF and Universal Sentence Encoder. Our results showed that SBERT performed the worst for the duration of 6 secs per question but the best for accuracy and fine-tuning on our course data improved the result.

Memory-based meta-learning is a powerful technique to build agents that adapt fast to any task within a target distribution. A previous theoretical study has argued that this remarkable performance is because the meta-training protocol incentivises agents to behave Bayes-optimally. We empirically investigate this claim on a number of prediction and bandit tasks. Inspired by ideas from theoretical computer science, we show that meta-learned and Bayes-optimal agents not only behave alike, but they even share a similar computational structure, in the sense that one agent system can approximately simulate the other. Furthermore, we show that Bayes-optimal agents are fixed points of the meta-learning dynamics. Our results suggest that memory-based meta-learning might serve as a general technique for numerically approximating Bayes-optimal agents - that is, even for task distributions for which we currently don't possess tractable models.

Spoken conversational question answering (SCQA) requires machines to model complex dialogue flow given the speech utterances and text corpora. Different from traditional text question answering (QA) tasks, SCQA involves audio signal processing, passage comprehension, and contextual understanding. However, ASR systems introduce unexpected noisy signals to the transcriptions, which result in performance degradation on SCQA. To overcome the problem, we propose CADNet, a novel contextualized attention-based distillation approach, which applies both cross-attention and self-attention to obtain ASR-robust contextualized embedding representations of the passage and dialogue history for performance improvements. We also introduce the spoken conventional knowledge distillation framework to distill the ASR-robust knowledge from the estimated probabilities of the teacher model to the student. We conduct extensive experiments on the Spoken-CoQA dataset and demonstrate that our approach achieves remarkable performance in this task.

In this paper, we study the possibility of almost unsupervised Multiple Choices Question Answering (MCQA). Starting from very basic knowledge, MCQA model knows that some choices have higher probabilities of being correct than the others. The information, though very noisy, guides the training of an MCQA model. The proposed method is shown to outperform the baseline approaches on RACE and even comparable with some supervised learning approaches on MC500.

With the rise of machines to human-level performance in complex recognition tasks, a growing amount of work is directed towards comparing information processing in humans and machines. These studies are an exciting chance to learn about one system by studying the other. Here, we propose ideas on how to design, conduct and interpret experiments such that they adequately support the investigation of mechanisms when comparing human and machine perception. We demonstrate and apply these ideas through three case studies. The first case study shows how human bias can affect how we interpret results, and that several analytic tools can help to overcome this human reference point. In the second case study, we highlight the difference between necessary and sufficient mechanisms in visual reasoning tasks. Thereby, we show that contrary to previous suggestions, feedback mechanisms might not be necessary for the tasks in question. The third case study highlights the importance of aligning experimental conditions. We find that a previously-observed difference in object recognition does not hold when adapting the experiment to make conditions more equitable between humans and machines. In presenting a checklist for comparative studies of visual reasoning in humans and machines, we hope to highlight how to overcome potential pitfalls in design or inference.

The authority that administers A-Level college entrance exams in the UK, Ofqual, recently found itself mired in scandal. Unable to hold live exams because of Covid-19, it designed and employed an algorithm that based scores partly on the historical performance of the schools students attended. The outcry was immediate, as students who were already disadvantaged found themselves further penalized by artificially deflated scores, their efforts disregarded and their futures thrown into disarray. This is far from an isolated incident. Even the world's most sophisticated technology companies have faced similar problems.