Unlearning algorithms aim to remove deleted data's influence from trained models at a cost lower than full retraining. However, prior guarantees of unlearning in literature are flawed and don't protect the privacy of deleted records. We show that when users delete their data as a function of published models, records in a database become interdependent. So, even retraining a fresh model after deletion of a record doesn't ensure its privacy. Secondly, unlearning algorithms that cache partial computations to speed up the processing can leak deleted information over a series of releases, violating the privacy of deleted records in the long run. To address these, we propose a sound deletion guarantee and show that the privacy of existing records is necessary for the privacy of deleted records. Under this notion, we propose an accurate, computationally efficient, and secure machine unlearning algorithm based on noisy gradient descent.

Models for text-to-image synthesis, such as DALL-E~2 and Stable Diffusion, have recently drawn a lot of interest from academia and the general public. These models are capable of producing high-quality images that depict a variety of concepts and styles when conditioned on textual descriptions. However, these models adopt cultural characteristics associated with specific Unicode scripts from their vast amount of training data, which may not be immediately apparent. We show that by simply inserting single non-Latin characters in a textual description, common models reflect cultural stereotypes and biases in their generated images. We analyze this behavior both qualitatively and quantitatively, and identify a model's text encoder as the root cause of the phenomenon. Additionally, malicious users or service providers may try to intentionally bias the image generation to create racist stereotypes by replacing Latin characters with similarly-looking characters from non-Latin scripts, so-called homoglyphs. To mitigate such unnoticed script attacks, we propose a novel homoglyph unlearning method to fine-tune a text encoder, making it robust against homoglyph manipulations.

The presence of decision-making algorithms in society is rapidly increasing nowadays, while concerns about their transparency and the possibility of these algorithms becoming new sources of discrimination are arising. There is a certain consensus about the need to develop AI applications with a Human-Centric approach. Human-Centric Machine Learning needs to be developed based on four main requirements: (i) utility and social good; (ii) privacy and data ownership; (iii) transparency and accountability; and (iv) fairness in AI-driven decision-making processes. All these four Human-Centric requirements are closely related to each other. With the aim of studying how current multimodal algorithms based on heterogeneous sources of information are affected by sensitive elements and inner biases in the data, we propose a fictitious case study focused on automated recruitment: FairCVtest. We train automatic recruitment algorithms using a set of multimodal synthetic profiles including image, text, and structured data, which are consciously scored with gender and racial biases. FairCVtest shows the capacity of the Artificial Intelligence (AI) behind automatic recruitment tools built this way (a common practice in many other application scenarios beyond recruitment) to extract sensitive information from unstructured data and exploit it in combination to data biases in undesirable (unfair) ways. We present an overview of recent works developing techniques capable of removing sensitive information and biases from the decision-making process of deep learning architectures, as well as commonly used databases for fairness research in AI. We demonstrate how learning approaches developed to guarantee privacy in latent spaces can lead to unbiased and fair automatic decision-making process.

In hindsight, ChatGPT may come to be seen as the greatest publicity stunt in AI history, an intoxicating glimpse at a future that may actually take years to realize--kind of like a 2012-vintage driverless car demo, but this time with a foretaste of an ethical guardrail that will take years to perfect. What ChatGPT delivered, in spades, that its predecessors like Microsoft Tay (released March 23, 2016, withdrawn March 24 for toxic behavior) and Meta's Galactica (released November 16, 2022, withdrawn November 18) could not, was an illusion--a sense that the problem of toxic spew was finally coming under control. ChatGPT rarely says anything overtly racist. Simple requests for anti-semitism and outright lies are often rebuffed. Indeed, at times it can seem so politically correct that the right wing has become enraged.

Potential obstacles to obtaining patent protection in Europe for an improvement in a general method for machine learning have been highlighted by a recent decision (T0702/20) from the EPO Board of Appeal. The decision relates to an application for a novel neural network apparatus having "loose coupling", based on an error code check matrix, between nodes of the neural network resulting in an initial configuration of the neural network that was argued to speed up training and operation of the apparatus while maintaining discrimination performance. The differences of the claimed invention over the prior art had been acknowledged during prosecution, but the Examining Division had rejected the Application on the basis that the distinguishing features "do not serve a technical purpose, and they are not related to a specific technical implementation either. They merely pertain to the initial, fixed structural definition of an abstract mathematical neural network-like model". During the Appeal, the Applicant provided several arguments as to why the claimed system did indeed serve a technical purpose which were not found persuasive by the Board. In response, the Board noted that a neural network can, in principle (if difficult in practice), be analysed to replace the inputs to each neuron by mathematical functions implemented by the nodes of the previous layer, and ultimately to obtain a mathematical description that describes the output of the neural network as a function of the input.

In this paper, we study the problem of legal domain adaptation problem from an imbalanced source domain to a partial target domain. The task aims to improve legal judgment predictions for non-professional fact descriptions. We formulate this task as a partial-and-imbalanced domain adaptation problem. Though deep domain adaptation has achieved cutting-edge performance in many unsupervised domain adaptation tasks. However, due to the negative transfer of samples in non-shared classes, it is hard for current domain adaptation model to solve the partial-and-imbalanced transfer problem. In this work, we explore large-scale non-shared but related classes data in the source domain with a hierarchy weighting adaptation to tackle this limitation. We propose to embed a novel pArtial Imbalanced Domain Adaptation technique (AIDA) in the deep learning model, which can jointly borrow sibling knowledge from non-shared classes to shared classes in the source domain and further transfer the shared classes knowledge from the source domain to the target domain. Experimental results show that our model outperforms the state-of-the-art algorithms.

AI-driven decision-making can lead to discrimination against certain individuals or social groups based on protected characteristics/attributes such as race, gender, or age. The domain of fairness-aware machine learning focuses on methods and algorithms for understanding, mitigating, and accounting for bias in AI/ML models. Still, thus far, the vast majority of the proposed methods assess fairness based on a single protected attribute, e.g. only gender or race. In reality, though, human identities are multi-dimensional, and discrimination can occur based on more than one protected characteristic, leading to the so-called ``multi-dimensional discrimination'' or ``multi-dimensional fairness'' problem. While well-elaborated in legal literature, the multi-dimensionality of discrimination is less explored in the machine learning community. Recent approaches in this direction mainly follow the so-called intersectional fairness definition from the legal domain, whereas other notions like additive and sequential discrimination are less studied or not considered thus far. In this work, we overview the different definitions of multi-dimensional discrimination/fairness in the legal domain as well as how they have been transferred/ operationalized (if) in the fairness-aware machine learning domain. By juxtaposing these two domains, we draw the connections, identify the limitations, and point out open research directions.

In this work, we combine the two paradigms: Federated Learning (FL) and Continual Learning (CL) for text classification task in cloud-edge continuum. The objective of Federated Continual Learning (FCL) is to improve deep learning models over life time at each client by (relevant and efficient) knowledge transfer without sharing data. Here, we address challenges in minimizing inter-client interference while knowledge sharing due to heterogeneous tasks across clients in FCL setup. In doing so, we propose a novel framework, Federated Selective Inter-client Transfer (FedSeIT) which selectively combines model parameters of foreign clients. To further maximize knowledge transfer, we assess domain overlap and select informative tasks from the sequence of historical tasks at each foreign client while preserving privacy. Evaluating against the baselines, we show improved performance, a gain of (average) 12.4\% in text classification over a sequence of tasks using five datasets from diverse domains. To the best of our knowledge, this is the first work that applies FCL to NLP.

Hazard and operability analysis (HAZOP) is the paradigm of industrial safety that can reveal the hazards of process from its node deviations, consequences, causes, measures and suggestions, and such hazards can be considered as hazard events (HaE). The classification research on HaE has much irreplaceable pragmatic values. In this paper, we present a novel deep learning model termed DLF through multifractal to explore HaE classification where the motivation is that HaE can be naturally regarded as a kind of time series. Specifically, first HaE is vectorized to get HaE time series by employing BERT. Then, a new multifractal analysis method termed HmF-DFA is proposed to win HaE fractal series by analyzing HaE time series. Finally, a new hierarchical gating neural network (HGNN) is designed to process HaE fractal series to accomplish the classification of HaE from three aspects: severity, possibility and risk. We take HAZOP reports of 18 processes as cases, and launch the experiments on this basis. Results demonstrate that compared with other classifiers, DLF classifier performs better under metrics of precision, recall and F1-score, especially for the severity aspect. Also, HmF-DFA and HGNN effectively promote HaE classification. Our HaE classification system can serve application incentives to experts, engineers, employees, and other enterprises. We hope our research can contribute added support to the daily practice in industrial safety.

The majority of existing post-hoc explanation approaches for machine learning models produce independent, per-variable feature attribution scores, ignoring a critical inherent characteristics of homogeneously structured data, such as visual or text data: there exist latent inter-variable relationships between features. In response, we develop a novel model-agnostic and permutation-based feature attribution approach based on the relational analysis between input variables. As a result, we are able to gain a broader insight into the predictions and decisions of machine learning models. Experimental evaluations of our framework in comparison with state-of-the-art attribution techniques on various setups involving both image and text data modalities demonstrate the effectiveness and validity of our method.