Optimal learning paths designed for E-learning systems using reinforcement learning.
problem Designing optimal learning paths for E-learning systems.
method Developed a hierarchical skill model and a proficiency level model, applied reinforcement learning to find the optimal learning strategy.
result Demonstrated the effectiveness of the proposed framework via numerical experiments.
A new method improves efficiency in finding optimal personalized treatment rules.
problem Heteroscedasticity and misspecified treatment-free effect models affect optimal ITR estimation.
method E-Learning framework that accounts for covariate-treatment dependent variance of residuals.
result E-Learning framework improves efficiency of optimal ITR estimation.
Study improves engagement prediction in educational videos.
problem Addressing cold-start problem in educational recommenders.
method Introduced VLE dataset with content and engagement features, conducted experiments.
result VLE dataset leads to better engagement prediction models.
A dataset for evaluating engagement with scientific video lectures.
problem Challenges in managing learning resources due to rapid creation of video lectures.
method Introduction of VLEngagement dataset with content-based and video-specific features, and metrics related to user engagement.
result The largest and most diverse publicly available dataset for understanding context-agnostic engagement in video lectures.
R2DE assesses new exam questions quickly and accurately.
problem Assessing newly generated exam questions efficiently and accurately.
method R2DE uses NLP to estimate difficulty and discrimination of questions based on their text.
result R2DE can estimate question difficulty and discrimination quickly and accurately.
EduQG generates better educational questions by pre-training on scientific text.
problem Improving the quality of educational questions for scalable self-assessment.
method Adapting a large language model for educational question generation, pre-trained on scientific text.
result EduQG produces superior educational questions compared to baseline models.
Deep learning improves evolutionary algorithms' adaptability.
problem Improving evolutionary algorithms' adaptability to various circumstances.
method Using deep reinforcement learning to dynamically adjust evolutionary algorithms' strategies.
result Deep learning enhances evolutionary algorithms' fitness increase and attainable fitness.
Deep neural networks have been developed drawing inspiration from the brain visual pathway, implementing an end-to-end approach: from image data to video object classes. However building an fMRI decoder with the typical structure of Convolutional Neural Network (CNN), i.e. learning multiple level of representations, se…
This work trains a model to generate all valid commands for text-based games.
problem Generating valid commands for text-based games.
method Training generative models on a dataset of text-based game contexts.
result The best model can generate valid commands unseen at training and achieves high F1 score.
Fourier PCA is Principal Component Analysis of a matrix obtained from higher order derivatives of the logarithm of the Fourier transform of a distribution.We make this method algorithmic by developing a tensor decomposition method for a pair of tensors sharing the same vectors in rank-1 decompositions. Our main appli…
The Gaussian process state space model (GPSSM) is a non-linear dynamical system, where unknown transition and/or measurement mappings are described by GPs. Most research in GPSSMs has focussed on the state estimation problem, i.e., computing a posterior of the latent state given the model. However, the key challenge in…
Arriving at the complete probabilistic knowledge of a domain, i.e., learning how all variables interact, is indeed a demanding task. In reality, settings often arise for which an individual merely possesses partial knowledge of the domain, and yet, is expected to give adequate answers to a variety of posed queries. Tha…
Develops a robust, fast, and widely-applicable neural architecture search method.
problem Inability of current NAS methods to be easily applied to new problems.
method Adaptive stochastic natural gradient method for simultaneous optimization of weights and architecture.
result Near state-of-the-art performances with low computational budgets.
This paper analyzes the stability and generalization of triplet learning algorithms.
problem Lack of theoretical understanding of triplet learning's generalization performance.
method Stability analysis and high-probability generalization bounds for triplet learning algorithms.
result Established general high-probability generalization bound for triplet learning algorithms.
A standard belief on emerging collective behavior is that it emerges from simple individual rules. Most of the mathematical research on such collective behavior starts from imperative individual rules, like always go to the center. But how could an (optimal) individual rule emerge during a short period within the group…
Paper proposes a method to learn multiple tasks without forgetting, maintaining model compactness.
problem Lifelong learning in deep learning models, especially forgetting of previous tasks.
method Combines deep model compression, critical weights selection, and progressive network expansion in an iterative manner.
result Incremental learning without forgetting, maintaining model compactness.
Machine learning discovers equations from simulated data.
problem Discovering equations from computer-generated data.
method Sparse regression for equation learning.
result Machine learning can discover equations from complex data.
Novel approach for robust domain generalization in health studies.
problem Challenges in making statistical inferences about underrepresented minority groups.
method Structured tensor completion for multi-dimensional domain generalization in linear regression models.
result Established rigorous theoretical guarantees and demonstrated minimax optimality.
Nowadays, the major challenge in machine learning is the Big Data challenge. The big data problems due to large number of data points or large number of features in each data point, or both, the training of models have become very slow. The training time has two major components: Time to access the data and time to pro…
TossingBot learns to throw objects accurately with residual physics.
problem Learning to throw arbitrary objects accurately and quickly.
method End-to-end formulation that learns control parameters from visual observations.
result TossingBot achieves 600+ grasps per hour with 85% throwing accuracy.
Model extraction attacks exploit MLaaS query interfaces to steal models.
problem Model extraction attacks compromise MLaaS systems by stealing models through query interfaces.
method Formalized model extraction, compared it to active learning, and investigated defense strategies.
result Recent active learning advancements can be used to implement powerful model extraction attacks.
The paper argues for interpreting neural networks as approximating the true posterior, enhancing in-context learning.
problem The limitations of traditional MLE interpretation in large-scale, single-epoch training setups.
method Demonstrates the power of interpreting neural networks as approximations of the true posterior, using experiments to predict generalizations.
result Models become robust in-context learners by effectively composing knowledge from their training data, revealing surprising generalizations.
Self-PU combines self-training with PU learning for improved binary classification.
problem Binary classification from unlabeled data and few labeled positives.
method Integrates self-paced training, self-calibrated loss, and self-distillation.
result Significantly improved performance on Alzheimer's Disease classification.
E&E uses contrastive learning to speed up SBI for high-dimensional systems.
problem Challenges in training high-dimensional emulators for complex systems.
method Contrastive learning for low-dimensional latent embedding and fast emulator.
result Superior performance in non-identifiable parameter estimation tasks.
Paper improves model accuracy with self-supervised learning.
problem Improving model accuracy with unlabeled data.
method Trains model on both original and self-supervised labels via input transformations.
result Significant accuracy improvement on fully-labeled datasets.
We propose channel charting (CC), a novel framework in which a multi-antenna network element learns a chart of the radio geometry in its surrounding area. The channel chart captures the local spatial geometry of the area so that points that are close in space will also be close in the channel chart and vice versa. CC w…
This paper explores the complexity of learning representations in contextual linear bandits.
problem Understanding the complexity of representation learning in contextual linear bandits.
method Systematic approach to representation learning in contextual linear bandits, focusing on instance-dependent perspective.
result Representation learning is fundamentally more complex than linear bandits, with some cases being arbitrarily harder.
Generative deep learning improves precipitation forecasts by adding resolution.
problem Inaccurate and unreliable precipitation forecasts due to unresolved processes.
method Applying GANs to super-resolve low-resolution weather model data using radar measurements.
result GANs and VAE-GANs produce high-resolution precipitation maps with better statistical properties than existing methods.
This paper studies node embeddings of networks, revealing their geometric properties.
problem Understanding the geometric properties of node embeddings in random networks.
method Characterization of ergodic limits, generalization, and convex relaxations of random walk node embedding objectives.
result The optimal node embedding Grammians have rank 1 for a nuclear norm relaxation of the non-randomized objective.
New learning rate schedule improves deep learning performance.
problem Improving deep learning performance with varying learning rates.
method Exponential learning rate schedule with Batch Normalization.
result Exponential learning rate schedule with BN is equivalent to standard BN + SGD + Weight Decay + Momentum.
Method improves few-shot one-class classification.
problem Learning binary classifier with data from only one class.
method Modified MAML algorithm to learn initialization for few-shot OCC.
result Method leads to better results than classical approaches.
New technique reduces imitation learning performance gap in finite samples.
problem Imitation learning performance gap in finite samples.
method Replay estimation to reduce empirical variance in finite samples.
result Achieves optimal performance gap of $\widetilde{O} \left( \min({H^{3/2}} / {N}, {H} / {\sqrt{N}}
ight)$.
Transformative machine learning improves model accuracy and explainability with limited data.
problem Improving model accuracy and interpretability with limited data in scientific tasks.
method Transforming intrinsic data representations to extrinsic ones based on model predictions.
result Transformative machine learning significantly outperforms intrinsic representations in drug-design, gene expression prediction, and meta-learning.
Efficiently updates vertex representations for dynamic graphs using random walks.
problem Updating vertex representations for dynamic graphs without re-generating them on each update.
method Proposes algorithms that extend random walk-based methods to dynamic graphs, considering the extent and rate of changes.
result Achieves competitive results to state-of-the-art methods while being computationally efficient.
We consider the task of low-multilinear-rank functional regression, i.e., learning a low-rank parametric representation of functions from scattered real-valued data. Our first contribution is the development and analysis of an efficient gradient computation that enables gradient-based optimization procedures, including…
Recursive causal discovery reduces errors and complexity in causal graph learning.
problem Challenges in causal discovery from limited data and computational complexity.
method Removable variables for recursive causal discovery, reducing problem size and CI tests.
result Worst-case performances nearly match lower bound, with state-of-the-art efficiency.
Proposes a method to learn fair data representations without requiring fair target data.
problem Achieving fairness in machine learning models, especially for human outcomes.
method Data-to-data translation, learning a mapping from input to a fair target domain using residuals and protected characteristics.
result Enforces fairness (equality of opportunity) by adjusting specific regions in face images and tabular data.
New algorithm for faster feature enumeration over finite fields.
problem Learning k-juntas over finite fields for multi-labeled data.
method Exploits Fourier detection techniques to develop an O(n^0.8k)-time algorithm.
result First non-trivial algorithm for k-juntas over F_q, answering Mossel et al.'s question.
Improvement-aware algorithms can achieve zero error in learning tasks.
problem Learning with agents who can improve their performance.
method Develops algorithms that account for agent improvement to achieve zero error.
result Improvement can reduce sample complexity or make learning harder.