Inferring unbiased treatment effects has received widespread attention in the machine learning community. In recent years, our community has proposed numerous solutions in standard settings, high-dimensional treatment settings, and even longitudinal settings. While very diverse, the solution has mostly relied on neural networks for inference and simultaneous correction of assignment bias. New approaches typically build on top of previous approaches by proposing new (or refined) architectures and learning algorithms. However, the end result -- a neural-network-based inference machine -- remains unchallenged. In this paper, we introduce a different type of solution in the longitudinal setting: a closed-form ordinary differential equation (ODE). While we still rely on continuous optimization to learn an ODE, the resulting inference machine is no longer a neural network. Doing so yields several advantages such as interpretability, irregular sampling, and a different set of identification assumptions. Above all, we consider the introduction of a completely new type of solution to be our most important contribution as it may spark entirely new innovations in treatment effects in general. We facilitate this by formulating our contribution as a framework that can transform any ODE discovery method into a treatment effects method.

Knowledge Measures (KMs) aim at quantifying the amount of knowledge/information that a knowledge base carries. On the other hand, Belief Change (BC) is the process of changing beliefs (in our case, in terms of contraction, expansion and revision) taking into account a new piece of knowledge, which possibly may be in contradiction with the current belief. We propose a new quantitative BC framework that is based on KMs by defining belief change operators that try to minimise, from an information-theoretic point of view, the surprise that the changed belief carries. To this end, we introduce the principle of minimal surprise. In particular, our contributions are (i) a general information-theoretic approach to KMs for which [1] is a special case; (ii) KM-based BC operators that satisfy the so-called AGM postulates; and (iii) a characterisation of any BC operator that satisfies the AGM postulates as a KM-based BC operator, i.e., any BC operator satisfying the AGM postulates can be encoded within our quantitative BC framework. We also introduce quantitative measures that account for the information loss of contraction, information gain of expansion and information change of revision. We also give a succinct look into the problem of iterated revision, which deals with the application of a sequence of revision operations in our framework, and also illustrate how one may build from our KM-based contraction operator also one not satisfying the (in)famous recovery postulate, by focusing on the so-called severe withdrawal model as an illustrative example.

Gender bias represents a form of systematic negative treatment that targets individuals based on their gender. This discrimination can range from subtle sexist remarks and gendered stereotypes to outright hate speech. Prior research has revealed that ignoring online abuse not only affects the individuals targeted but also has broader societal implications. These consequences extend to the discouragement of women's engagement and visibility within public spheres, thereby reinforcing gender inequality. This thesis investigates the nuances of how gender bias is expressed through language and within language technologies. Significantly, this thesis expands research on gender bias to multilingual contexts, emphasising the importance of a multilingual and multicultural perspective in understanding societal biases. In this thesis, I adopt an interdisciplinary approach, bridging natural language processing with other disciplines such as political science and history, to probe gender bias in natural language and language models.

Video-based surgical instrument segmentation plays an important role in robot-assisted surgeries. Unlike supervised settings, unsupervised segmentation relies heavily on motion cues, which are challenging to discern due to the typically lower quality of optical flow in surgical footage compared to natural scenes. This presents a considerable burden for the advancement of unsupervised segmentation techniques. In our work, we address the challenge of enhancing model performance despite the inherent limitations of low-quality optical flow. Our methodology employs a three-pronged approach: extracting boundaries directly from the optical flow, selectively discarding frames with inferior flow quality, and employing a fine-tuning process with variable frame rates. We thoroughly evaluate our strategy on the EndoVis2017 VOS dataset and Endovis2017 Challenge dataset, where our model demonstrates promising results, achieving a mean Intersection-over-Union (mIoU) of 0.75 and 0.72, respectively. Our findings suggest that our approach can greatly decrease the need for manual annotations in clinical environments and may facilitate the annotation process for new datasets. The code is available at https://github.com/wpr1018001/Rethinking-Low-quality-Optical-Flow.git

Reinforcement Learning from Human Feedback (RLHF) has proven to be a strong method to align Pretrained Large Language Models (LLMs) with human preferences. But training models with RLHF is computationally expensive, and an overall complex process. In this work, we study RLHF where the underlying models are trained using the parameter efficient method of Low-Rank Adaptation (LoRA) introduced by Hu et al. [2021]. We investigate the setup of "Parameter Efficient Reinforcement Learning" (PERL), in which we perform reward model training and reinforcement learning using LoRA. We compare PERL to conventional fine-tuning (full-tuning) across various configurations for 7 benchmarks, including 2 novel datasets, of reward modeling and reinforcement learning. We find that PERL performs on par with the conventional RLHF setting, while training faster, and with less memory. This enables the high performance of RLHF, while reducing the computational burden that limits its adoption as an alignment technique for Large Language Models. We also release 2 novel thumbs up/down preference datasets: "Taskmaster Coffee", and "Taskmaster Ticketing" to promote research around RLHF.

Language models (LMs) can express factual knowledge involving numeric properties such as Karl Popper was born in 1902. However, how this information is encoded in the model's internal representations is not understood well. Here, we introduce a simple method for finding and editing representations of numeric properties such as an entity's birth year. Empirically, we find low-dimensional subspaces that encode numeric properties monotonically, in an interpretable and editable fashion. When editing representations along directions in these subspaces, LM output changes accordingly. For example, by patching activations along a "birthyear" direction we can make the LM express an increasingly late birthyear: Karl Popper was born in 1929, Karl Popper was born in 1957, Karl Popper was born in 1968. Property-encoding directions exist across several numeric properties in all models under consideration, suggesting the possibility that monotonic representation of numeric properties consistently emerges during LM pretraining. Code: https://github.com/bheinzerling/numeric-property-repr

Large language models (LLMs) have shown excellent performance on various NLP tasks. To use LLMs as strong sequential recommenders, we explore the in-context learning approach to sequential recommendation. We investigate the effects of instruction format, task consistency, demonstration selection, and number of demonstrations. As increasing the number of demonstrations in ICL does not improve accuracy despite using a long prompt, we propose a novel method called LLMSRec-Syn that incorporates multiple demonstration users into one aggregated demonstration. Our experiments on three recommendation datasets show that LLMSRec-Syn outperforms state-of-the-art LLM-based sequential recommendation methods. In some cases, LLMSRec-Syn can perform on par with or even better than supervised learning methods.

Station-keeping tasks for high-altitude balloons show promise in areas such as ecological surveys, atmospheric analysis, and communication relays. However, identifying the optimal time and position to launch a latex high-altitude balloon is still a challenging and multifaceted problem. For example, tasks such as forest fire tracking place geometric constraints on the launch location of the balloon. Furthermore, identifying the most optimal location also heavily depends on atmospheric conditions. We first illustrate how reinforcement learning-based controllers, frequently used for station-keeping tasks, can exploit the environment. This exploitation can degrade performance on unseen weather patterns and affect station-keeping performance when identifying an optimal launch configuration. Valuing all states equally in the region, the agent exploits the region's geometry by flying near the edge, leading to risky behaviours. We propose a modification which compensates for this exploitation and finds this leads to, on average, higher steps within the target region on unseen data. Then, we illustrate how Bayesian Optimisation (BO) can identify the optimal launch location to perform station-keeping tasks, maximising the expected undiscounted return from a given rollout. We show BO can find this launch location in fewer steps compared to other optimisation methods. Results indicate that, surprisingly, the most optimal location to launch from is not commonly within the target region. Please find further information about our project at https://sites.google.com/view/bo-lauch-balloon/.

In recent years, epidemic policy-making models are increasingly being used to provide reference for governors on prevention and control policies against catastrophic epidemics such as SARS, H1N1 and COVID-19. Existing studies are currently constrained by two issues: First, previous methods develop policies based on effect evaluation, since few of factors in real-world decision-making can be modeled, the output policies will then easily become extreme. Second, the subjectivity and cognitive limitation of human make the historical policies not always optimal for the training of decision models. To these ends, we present a novel Policy Combination Synthesis (PCS) model for epidemic policy-making. Specially, to prevent extreme decisions, we introduce adversarial learning between the model-made policies and the real policies to force the output policies to be more human-liked. On the other hand, to minimize the impact of sub-optimal historical policies, we employ contrastive learning to let the model draw on experience from the best historical policies under similar scenarios. Both adversarial and contrastive learning are adaptive based on the comprehensive effects of real policies to ensure the model always learns useful information. Extensive experiments on real-world data prove the effectiveness of the proposed model.

We present a knowledge integration framework (called KIF) that uses Wikidata as a lingua franca to integrate heterogeneous knowledge bases. These can be triplestores, relational databases, CSV files, etc., which may or may not use the Wikidata dialect of RDF. KIF leverages Wikidata's data model and vocabulary plus user-defined mappings to expose a unified view of the integrated bases while keeping track of the context and provenance of their statements. The result is a virtual knowledge base which behaves like an "extended Wikidata" and which can be queried either through an efficient filter interface or using SPARQL. We present the design and implementation of KIF, discuss how we have used it to solve a real integration problem in the domain of chemistry (involving Wikidata, PubChem, and IBM CIRCA), and present experimental results on the performance and overhead of KIF.