In this study, we propose a simple and effective preprocessing method for subword segmentation based on a data compression algorithm. Compression-based subword segmentation has recently attracted significant attention as a preprocessing method for training data in Neural Machine Translation. Among them, BPE/BPE-dropout is one of the fastest and most effective method compared to conventional approaches. However, compression-based approach has a drawback in that generating multiple segmentations is difficult due to the determinism. To overcome this difficulty, we focus on a probabilistic string algorithm, called locally-consistent parsing (LCP), that has been applied to achieve optimum compression. Employing the probabilistic mechanism of LCP, we propose LCP-dropout for multiple subword segmentation that improves BPE/BPE-dropout, and show that it outperforms various baselines in learning from especially small training data.

This paper stresses the importance of biases in the field of artificial intelligence (AI) in two regards. First, in order to foster efficient algorithmic decision-making in complex, unstable, and uncertain real-world environments, we argue for the structurewise implementation of human cognitive biases in learning algorithms. Secondly, we argue that in order to achieve ethical machine behavior, filter mechanisms have to be applied for selecting biased training stimuli that represent social or behavioral traits that are ethically desirable. We use insights from cognitive science as well as ethics and apply them to the AI field, combining theoretical considerations with seven case studies depicting tangible bias implementation scenarios. Ultimately, this paper is the first tentative step to explicitly pursue the idea of a re-evaluation of the ethical significance of machine biases, as well as putting the idea forth to implement cognitive biases into machines.

Everyone wants to write beautiful and correct text, yet the lack of language skills, experience, or hasty typing can result in errors. By employing the recent advances in transformer architectures, we construct a grammatical error correction model for Lithuanian, the language rich in archaic features. We compare subword and byte-level approaches and share our best trained model, achieving F$_{0.5}$=0.92, and accompanying code, in an online open-source repository.

Expectation-Maximization (EM) algorithm is a popular choice for learning latent variable models. Variants of the EM have been initially introduced by Neal and Hinton (1998), using incremental updates to scale to large datasets, and by Wei and Tanner (1990); Delyon et al. (1999), using Monte Carlo (MC) approximations to bypass the intractable conditional expectation of the latent data for most nonconvex models. In this paper, we propose a general class of methods called Two-Timescale EM Methods based on a two-stage approach of stochastic updates to tackle an essential nonconvex optimization task for latent variable models. We motivate the choice of a double dynamic by invoking the variance reduction virtue of each stage of the method on both sources of noise: the index sampling for the incremental update and the MC approximation. We establish finite-time and global convergence bounds for nonconvex objective functions. Numerical applications on various models such as deformable template for image analysis or nonlinear models for pharmacokinetics are also presented to illustrate our findings.

In recent years, many works have investigated the feasibility of conversational robots for performing specific tasks, such as healthcare and interview. Along with this development comes a practical issue: how should we synthesize robotic voices to meet the needs of different situations? In this paper, we discuss this issue from three perspectives: 1) the difficulties of synthesizing non-verbal and interaction-oriented speech signals, particularly backchannels; 2) the scenario classification for robotic voice synthesis; 3) the ethical issues regarding the design of robot voice for its emotion and identity. We present the findings of relevant literature and our prior work, trying to bring the attention of human-robot interaction researchers to design better conversational robots in the future.

In this paper, we analyze the generalization performance of the Iterative Hard Thresholding (IHT) algorithm widely used for sparse recovery problems. The parameter estimation and sparsity recovery consistency of IHT has long been known in compressed sensing. From the perspective of statistical learning, another fundamental question is how well the IHT estimation would predict on unseen data. This paper makes progress towards answering this open question by introducing a novel sparse generalization theory for IHT under the notion of algorithmic stability. Our theory reveals that: 1) under natural conditions on the empirical risk function over $n$ samples of dimension $p$, IHT with sparsity level $k$ enjoys an $\mathcal{\tilde O}(n^{-1/2}\sqrt{k\log(n)\log(p)})$ rate of convergence in sparse excess risk; 2) a tighter $\mathcal{\tilde O}(n^{-1/2}\sqrt{\log(n)})$ bound can be established by imposing an additional iteration stability condition on a hypothetical IHT procedure invoked to the population risk; and 3) a fast rate of order $\mathcal{\tilde O}\left(n^{-1}k(\log^3(n)+\log(p))\right)$ can be derived for strongly convex risk function under proper strong-signal conditions. The results have been substantialized to sparse linear regression and sparse logistic regression models to demonstrate the applicability of our theory. Preliminary numerical evidence is provided to confirm our theoretical predictions.

Many algorithms, whether supervised or unsupervised, make use of distance measures. These measures, such as euclidean distance or cosine similarity, can often be found in algorithms such as k-NN, UMAP, HDBSCAN, etc. Understanding the field of distance measures is more important than you might realize. Take k-NN for example, a technique often used for supervised learning. As a default, it often uses euclidean distance. However, what if your data is highly dimensional?

Films and TV programmes have long portrayed caveman as using grunts to communicate with one another. But a new study suggests that our ancient ancestors likely did not use sounds to communicate, and instead opted for hand gestures. Researchers from the University of Western Australia asked volunteers to attempt to describe words using only grunts or gestures. They found that gestures were far more effective in communicating meaning and were often similar between cultures. 'The universality of gesture means it is ideally suited to bootstrapping human communication among modern humans and therefore supports the hypothesis that gesture is the primary modality for language creation,' the researchers said in their study, published in Proceedings of the Royal Society B. Films and TV programmes have long portrayed caveman as using grunts to communicate with one another. Searching for a way to make your point?

Conversational teachable agents offer a promising platform to support learning, both in the classroom and in remote settings. In this context, the agent takes the role of the novice, while the student takes on the role of teacher. This framing is significant for its ability to elicit the Prot\'eg\'e effect in the student-teacher, a pedagogical phenomenon known to increase engagement in the teaching task, and also improve cognitive outcomes. In prior work, teachable agents often take a passive role in the learning interaction, and there are few studies in which the agent and student engage in natural language dialogue during the teaching task. This work investigates the effect of teaching modality when interacting with a virtual agent, via the web-based teaching platform, the Curiosity Notebook. A method of teaching the agent by selecting sentences from source material is compared to a method paraphrasing the source material and typing text input to teach. A user study has been conducted to measure the effect teaching modality on the learning outcomes and engagement of the participants. The results indicate that teaching via paraphrasing and text input has a positive effect on learning outcomes for the material covered, and also on aspects of affective engagement. Furthermore, increased paraphrasing effort, as measured by the similarity between the source material and the material the teacher conveyed to the robot, improves learning outcomes for participants.

We propose a framework to modularize the training of neural language models that use diverse forms of sentence-external context (including metadata) by eliminating the need to jointly train sentence-external and within-sentence encoders. Our approach, contextual universal embeddings (CUE), trains LMs on one set of context, such as date and author, and adapts to novel metadata types, such as article title, or previous sentence. The model consists of a pretrained neural sentence LM, a BERT-based context encoder, and a masked transformer decoder that estimates LM probabilities using sentence-internal and sentence-external information. When context or metadata are unavailable, our model learns to combine contextual and sentence-internal information using noisy oracle unigram embeddings as a proxy. Real contextual information can be introduced later and used to adapt a small number of parameters that map contextual data into the decoder's embedding space. We validate the CUE framework on a NYTimes text corpus with multiple metadata types, for which the LM perplexity can be lowered from 36.6 to 27.4 by conditioning on context. Bootstrapping a contextual LM with only a subset of the context/metadata during training retains 85\% of the achievable gain. Training the model initially with proxy context retains 67% of the perplexity gain after adapting to real context. Furthermore, we can swap one type of pretrained sentence LM for another without retraining the context encoders, by only adapting the decoder model. Overall, we obtain a modular framework that allows incremental, scalable training of context-enhanced LMs.