Supervised learning algorithms generally assume the availability of enough memory to store their data model during the training and test phases. However, in the Internet of Things, this assumption is unrealistic when data comes in the form of infinite data streams, or when learning algorithms are deployed on devices with reduced amounts of memory. In this paper, we adapt the online Mondrian forest classification algorithm to work with memory constraints on data streams. In particular, we design five out-of-memory strategies to update Mondrian trees with new data points when the memory limit is reached. Moreover, we design trimming mechanisms to make Mondrian trees more robust to concept drifts under memory constraints. We evaluate our algorithms on a variety of real and simulated datasets, and we conclude with recommendations on their use in different situations: the Extend Node strategy appears as the best out-of-memory strategy in all configurations, whereas different trimming mechanisms should be adopted depending on whether a concept drift is expected. All our methods are implemented in the OrpailleCC open-source library and are ready to be used on embedded systems and connected objects.

We propose a scheme for supervised image classification that uses privileged information, in the form of keypoint annotations for the training data, to learn strong models from small and/or biased training sets. Our main motivation is the recognition of animal species for ecological applications such as biodiversity modelling, which is challenging because of long-tailed species distributions due to rare species, and strong dataset biases such as repetitive scene background in camera traps. To counteract these challenges, we propose a visual attention mechanism that is supervised via keypoint annotations that highlight important object parts. This privileged information, implemented as a novel privileged pooling operation, is only required during training and helps the model to focus on regions that are discriminative. In experiments with three different animal species datasets, we show that deep networks with privileged pooling can use small training sets more efficiently and generalize better.

Disordered many-body systems exhibit a wide range of emergent phenomena across different scales. These complex behaviors can be utilized for various information processing tasks such as error correction, learning, and optimization. Despite the empirical success of utilizing these systems for intelligent tasks, the underlying principles that govern their emergent intelligent behaviors remain largely unknown. In this thesis, we aim to characterize such emergent intelligence in disordered systems through statistical physics. We chart a roadmap for our efforts in this thesis based on two axes: learning mechanisms (long-term memory vs. working memory) and learning dynamics (artificial vs. natural). Throughout our journey, we uncover relationships between learning mechanisms and physical dynamics that could serve as guiding principles for designing intelligent systems. We hope that our investigation into the emergent intelligence of seemingly disparate learning systems can expand our current understanding of intelligence beyond neural systems and uncover a wider range of computational substrates suitable for AI applications.

This study investigates the consistency of feedback ratings generated by OpenAI's GPT-4, a state-of-the-art artificial intelligence language model, across multiple iterations, time spans and stylistic variations. The model rated responses to tasks within the Higher Education (HE) subject domain of macroeconomics in terms of their content and style. Statistical analysis was conducted in order to learn more about the interrater reliability, consistency of the ratings across iterations and the correlation between ratings in terms of content and style. The results revealed a high interrater reliability with ICC scores ranging between 0.94 and 0.99 for different timespans, suggesting that GPT-4 is capable of generating consistent ratings across repetitions with a clear prompt. Style and content ratings show a high correlation of 0.87. When applying a non-adequate style the average content ratings remained constant, while style ratings decreased, which indicates that the large language model (LLM) effectively distinguishes between these two criteria during evaluation. The prompt used in this study is furthermore presented and explained. Further research is necessary to assess the robustness and reliability of AI models in various use cases.

Second-order optimization algorithms exhibit excellent convergence properties for training deep learning models, but often incur significant computation and memory overheads. This can result in lower training efficiency than the first-order counterparts such as stochastic gradient descent (SGD). In this work, we present a memory- and time-efficient second-order algorithm named Eva with two novel techniques: 1) we construct the second-order information with the Kronecker factorization of small stochastic vectors over a mini-batch of training data to reduce memory consumption, and 2) we derive an efficient update formula without explicitly computing the inverse of matrices using the Sherman-Morrison formula. We further extend Eva to a general vectorized approximation framework to improve the compute and memory efficiency of two existing second-order algorithms (FOOF and Shampoo) without affecting their convergence performance. Extensive experimental results on different models and datasets show that Eva reduces the end-to-end training time up to 2.05x and 2.42x compared to first-order SGD and second-order algorithms (K-FAC and Shampoo), respectively.

Continual learning seeks to enable deep learners to train on a series of tasks of unknown length without suffering from the catastrophic forgetting of previous tasks. One effective solution is replay, which involves storing few previous experiences in memory and replaying them when learning the current task. However, there is still room for improvement when it comes to selecting the most informative samples for storage and determining the optimal number of samples to be stored. This study aims to address these issues with a novel comparison of the commonly used reservoir sampling to various alternative population strategies and providing a novel detailed analysis of how to find the optimal number of stored samples.

Fine-tuning language models in a downstream task is the standard approach for many state-of-the-art methodologies in the field of NLP. However, when the distribution between the source task and target task drifts, \textit{e.g.}, conversational environments, these gains tend to be diminished. This article proposes a sequence of pre-training steps (a curriculum) guided by "data hacking" and grammar analysis that allows further gradual adaptation between pre-training distributions. In our experiments, we acquire a considerable improvement from our method compared to other known pre-training approaches for the MultiWoZ task.

Multitask learning is a powerful framework that enables one to simultaneously learn multiple related tasks by sharing information between them. Quantifying uncertainty in the estimated tasks is of pivotal importance for many downstream applications, such as online or active learning. In this work, we provide novel multitask confidence intervals in the challenging agnostic setting, i.e., when neither the similarity between tasks nor the tasks' features are available to the learner. The obtained intervals do not require i.i.d. data and can be directly applied to bound the regret in online learning. Through a refined analysis of the multitask information gain, we obtain new regret guarantees that, depending on a task similarity parameter, can significantly improve over treating tasks independently. We further propose a novel online learning algorithm that achieves such improved regret without knowing this parameter in advance, i.e., automatically adapting to task similarity. As a second key application of our results, we introduce a novel multitask active learning setup where several tasks must be simultaneously optimized, but only one of them can be queried for feedback by the learner at each round. For this problem, we design a no-regret algorithm that uses our confidence intervals to decide which task should be queried. Finally, we empirically validate our bounds and algorithms on synthetic and real-world (drug discovery) data.

DOLCE, the first top-level (foundational) ontology to be axiomatized, has remained stable for twenty years and today is broadly used in a variety of domains. DOLCE is inspired by cognitive and linguistic considerations and aims to model a commonsense view of reality, like the one human beings exploit in everyday life in areas as diverse as socio-technical systems, manufacturing, financial transactions and cultural heritage. DOLCE clearly lists the ontological choices it is based upon, relies on philosophical principles, is richly formalized, and is built according to well-established ontological methodologies, e.g. OntoClean. Because of these features, it has inspired most of the existing top-level ontologies and has been used to develop or improve standards and public domain resources (e.g. CIDOC CRM, DBpedia and WordNet). Being a foundational ontology, DOLCE is not directly concerned with domain knowledge. Its purpose is to provide the general categories and relations needed to give a coherent view of reality, to integrate domain knowledge, and to mediate across domains. In these 20 years DOLCE has shown that applied ontologies can be stable and that interoperability across reference and domain ontologies is a reality. This paper briefly introduces the ontology and shows how to use it on a few modeling cases.

Unsupervised representation learning approaches aim to learn discriminative feature representations from unlabeled data, without the requirement of annotating every sample. Enabling unsupervised representation learning is extremely crucial for time series data, due to its unique annotation bottleneck caused by its complex characteristics and lack of visual cues compared with other data modalities. In recent years, unsupervised representation learning techniques have advanced rapidly in various domains. However, there is a lack of systematic analysis of unsupervised representation learning approaches for time series. To fill the gap, we conduct a comprehensive literature review of existing rapidly evolving unsupervised representation learning approaches for time series. Moreover, we also develop a unified and standardized library, named ULTS (i.e., Unsupervised Learning for Time Series), to facilitate fast implementations and unified evaluations on various models. With ULTS, we empirically evaluate state-of-the-art approaches, especially the rapidly evolving contrastive learning methods, on 9 diverse real-world datasets. We further discuss practical considerations as well as open research challenges on unsupervised representation learning for time series to facilitate future research in this field.