Topological data analysis (TDA) is a powerful technique for extracting complex and valuable shape-related summaries of high-dimensional data. However, the computational demands of classical algorithms for computing TDA are exorbitant, and quickly become impractical for high-order characteristics. Quantum computers offer the potential of achieving significant speedup for certain computational problems. Indeed, TDA has been purported to be one such problem, yet, quantum computing algorithms proposed for the problem, such as the original Quantum TDA (QTDA) formulation by Lloyd, Garnerone and Zanardi, require fault-tolerance qualifications that are currently unavailable. In this study, we present NISQ-TDA, a fully implemented end-to-end quantum machine learning algorithm needing only a short circuit-depth, that is applicable to high-dimensional classical data, and with provable asymptotic speedup for certain classes of problems. The algorithm neither suffers from the data-loading problem nor does it need to store the input data on the quantum computer explicitly. The algorithm was successfully executed on quantum computing devices, as well as on noisy quantum simulators, applied to small datasets. Preliminary empirical results suggest that the algorithm is robust to noise.

Parametrized machine learning models for inference often include non-linear and nonconvex constraints over the parameters and meta-parameters. Training these models to convergence is in general difficult, and naive methods such as projected gradient descent or grid search are not easily able to enforce the functional constraints. This work explores the optimization of a constrained neural network (familiar from machine learning but with parameter constraints), in the service of neuro-symbolic logical reasoning. Logical neural networks (LNNs) provide a well-justified, interpretable example of training under non-trivial constraints. In this paper, we propose a unified framework for solving this nonlinear programming problem by leveraging primal-dual optimization methods, and quantify the corresponding convergence rate to the Karush-Kuhn-Tucker (KKT) points of this problem. Extensive numerical results on both a toy example and training an LNN over real datasets validate the efficacy of the method.

Knowledge capture from human experts in domain-specific settings can benefit from incisive use of machine intelligence to reduce expended time and effort. Such a capability can be of significant value to deep learning, given its demand for large labeled data. We propose an ML-based system for interactive labeling of image datasets to speed up class attribution performed by domain experts. The tool visualizes feature spaces and makes it directly editable through online integration of applied labels. We propose realistic annotation emulation to evaluate the system design of interactive active learning, based on our improved semi-supervised extension of t-SNE dimensionality reduction. We contribute globally normalized attractions, semi-supervised repulsion, smoothed label integration, and parameter optimization in our improved t-SNE. Our active learning tool can significantly increase labeling efficiency compared to uncertainty sampling, and we show that less than 100 labeling actions are typically sufficient for good classification on a variety of specialized image datasets. Our contribution is unique given that it needs to perform dimensionality reduction, feature space visualization and editing, interactive label propagation, low-complexity active learning, human perceptual modeling, annotation emulation and unsupervised feature extraction for specialized datasets in a production-quality implementation.

In low-resource, over-burdened schools and learning centres, peer assessment systems promise significant practical and pedagogical benefits. Many of the these benefits have been realised in contexts like massive open online courses (MOOCs) and university classrooms which share a specific trait with low-resource schools: high learner-teacher ratios. However, the constraints and considerations for designing and deploying peer assessment systems in low-resource classrooms have not been well-researched and understood, especially for high school. In this paper, we present the design of a peer assessment system for second language learning (English as a Second Language) for high school learners in South Africa. We report findings from multiple studies investigating qualitative and quantitative aspects of peer review, as well as the contextual factors that influence the viability of peer assessment systems in these contexts.

Supervised deep learning depends on labeled datasets to define objective categorization of subject matter, but annotation is typically quite expensive for specialized domains. The ISIC 2017 skin lesion and BCCD blood cell image datasets are used to represent complex medical annotation scenarios, where domain knowledge is not permitted in either preprocessing or feature extraction. A low complexity supervision method is proposed, based on an iterative machine learning algorithm that fulfills the requirements for cognitive-assisted labeling. The visualization and editing of feature spaces is demanded where new label information must be integrated to improve the embedding quality as feedback mechanism. The annotators ability for fast local homogeneity assessment is leveraged through compound labeling prospects, which is the basis for achieving efficient labeling. Improved unsupervised feature extraction is hypothesized to reduce the labeling burden so the best feature extractors are empirically located at the various depths in ImageNet-pretrained convolutional neural networks, including VGG-16, Inception-v4 and Inception-Resnet-v2. Annotator emulation is performed to simulate upper bounds of achievable labeling efficiency and to explore active learning dynamics. A two-fold increase in efficiency is shown in the case of partial labeling, despite the complexity of the skin lesion data and the marginal improvement with pretrained features.