This report examines whether advanced AIs that perform well in training will be doing so in order to gain power later -- a behavior I call "scheming" (also sometimes called "deceptive alignment"). I conclude that scheming is a disturbingly plausible outcome of using baseline machine learning methods to train goal-directed AIs sophisticated enough to scheme (my subjective probability on such an outcome, given these conditions, is roughly 25%). In particular: if performing well in training is a good strategy for gaining power (as I think it might well be), then a very wide variety of goals would motivate scheming -- and hence, good training performance. This makes it plausible that training might either land on such a goal naturally and then reinforce it, or actively push a model's motivations towards such a goal as an easy way of improving performance. What's more, because schemers pretend to be aligned on tests designed to reveal their motivations, it may be quite difficult to tell whether this has occurred. However, I also think there are reasons for comfort. In particular: scheming may not actually be such a good strategy for gaining power; various selection pressures in training might work against schemer-like goals (for example, relative to non-schemers, schemers need to engage in extra instrumental reasoning, which might harm their training performance); and we may be able to increase such pressures intentionally. The report discusses these and a wide variety of other considerations in detail, and it suggests an array of empirical research directions for probing the topic further.

Federated learning is a new learning paradigm that decouples data collection and model training via multi-party computation and model aggregation. As a flexible learning setting, federated learning has the potential to integrate with other learning frameworks. We conduct a focused survey of federated learning in conjunction with other learning algorithms. Specifically, we explore various learning algorithms to improve the vanilla federated averaging algorithm and review model fusion methods such as adaptive aggregation, regularization, clustered methods, and Bayesian methods. Following the emerging trends, we also discuss federated learning in the intersection with other learning paradigms, termed federated X learning, where X includes multitask learning, meta-learning, transfer learning, unsupervised learning, and reinforcement learning. This survey reviews the state of the art, challenges, and future directions.

The ability to construct a realistic simulator of financial exchanges, including reproducing the dynamics of the limit order book, can give insight into many counterfactual scenarios, such as a flash crash, a margin call, or changes in macroeconomic outlook. In recent years, agent-based models have been developed that reproduce many features of an exchange, as summarised by a set of stylised facts and statistics. However, the ability to calibrate simulators to a specific period of trading remains an open challenge. In this work, we develop a novel approach to the calibration of market simulators by leveraging recent advances in deep learning, specifically using neural density estimators and embedding networks. We demonstrate that our approach is able to correctly identify high probability parameter sets, both when applied to synthetic and historical data, and without reliance on manually selected or weighted ensembles of stylised facts.

Controlling the departure time of the trucks from a container hub is important to both the traffic and the logistics systems. This, however, requires an intelligent decision support system that can control and manage truck arrival times at terminal gates. This paper introduces an integrated model that can be used to understand, predict, and control logistics and traffic interactions in the port-hinterland ecosystem. This approach is context-aware and makes use of big historical data to predict system states and apply control policies accordingly, on truck inflow and outflow. The control policies ensure multiple stakeholders satisfaction including those of trucking companies, terminal operators, and road traffic agencies. The proposed method consists of five integrated modules orchestrated to systematically steer truckers toward choosing those time slots that are expected to result in lower gate waiting times and more cost-effective schedules. The simulation is supported by real-world data and shows that significant gains can be obtained in the system.

This paper reports the overview of the VLSP 2022 - Vietnamese abstractive multi-document summarization (Abmusu) shared task for Vietnamese News. This task is hosted at the 9$^{th}$ annual workshop on Vietnamese Language and Speech Processing (VLSP 2022). The goal of Abmusu shared task is to develop summarization systems that could create abstractive summaries automatically for a set of documents on a topic. The model input is multiple news documents on the same topic, and the corresponding output is a related abstractive summary. In the scope of Abmusu shared task, we only focus on Vietnamese news summarization and build a human-annotated dataset of 1,839 documents in 600 clusters, collected from Vietnamese news in 8 categories. Participated models are evaluated and ranked in terms of \texttt{ROUGE2-F1} score, the most typical evaluation metric for document summarization problem.

Entities like person, location, organization are important for literary text analysis. The lack of annotated data hinders the progress of named entity recognition (NER) in literary domain. To promote the research of literary NER, we build the largest multi-genre literary NER corpus containing 263,135 entities in 105,851 sentences from 260 online Chinese novels spanning 13 different genres. Based on the corpus, we investigate characteristics of entities from different genres. We propose several baseline NER models and conduct cross-genre and cross-domain experiments. Experimental results show that genre difference significantly impact NER performance though not as much as domain difference like literary domain and news domain. Compared with NER in news domain, literary NER still needs much improvement and the Out-of-Vocabulary (OOV) problem is more challenging due to the high variety of entities in literary works.

The control design tools for linear systems typically involves pole placement and computing Lyapunov functions which are useful for ensuring stability. But given higher requirements on control design, a designer is expected to satisfy other specification such as safety or temporal logic specification as well, and a naive control design might not satisfy such specification. A control designer can employ model checking as a tool for checking safety and obtain a counterexample in case of a safety violation. While several scalable techniques for verification have been developed for safety verification of linear dynamical systems, such tools merely act as decision procedures to evaluate system safety and, consequently, yield a counterexample as an evidence to safety violation. However these model checking methods are not geared towards discovering corner cases or re-using verification artifacts for another sub-optimal safety specification. In this paper, we describe a technique for obtaining complete characterization of counterexamples for a safety violation in linear systems. The proposed technique uses the reachable set computed during safety verification for a given temporal logic formula, performs constraint propagation, and represents all modalities of counterexamples using a binary decision diagram (BDD). We introduce an approach to dynamically determine isomorphic nodes for obtaining a considerably reduced (in size) decision diagram. A thorough experimental evaluation on various benchmarks exhibits that the reduction technique achieves up to $67\%$ reduction in the number of nodes and $75\%$ reduction in the width of the decision diagram.

This paper presents a modeling approach to infer scheduling and routing patterns from digital freight transport activity data for different freight markets. We provide a complete modeling framework including a new discrete-continuous decision tree approach for extracting rules from the freight transport data. We apply these models to collected tour data for the Netherlands to understand departure time patterns and tour strategies, also allowing us to evaluate the effectiveness of the proposed algorithm. We find that spatial and temporal characteristics are important to capture the types of tours and time-of-day patterns of freight activities. Also, the empirical evidence indicates that carriers in most of the transport markets are sensitive to the level of congestion. Many of them adjust the type of tour, departure time, and the number of stops per tour when facing a congested zone. The results can be used by practitioners to get more grip on transport markets and develop freight and traffic management measures.

Optimization can be found in many real-life applications. Designing an effective algorithm for a specific optimization problem typically requires a tedious amount of effort from human experts with domain knowledge and algorithm design skills. In this paper, we propose a novel approach called Algorithm Evolution using Large Language Model (AEL). It utilizes a large language model (LLM) to automatically generate optimization algorithms via an evolutionary framework. AEL does algorithm-level evolution without model training. Human effort and requirements for domain knowledge can be significantly reduced. We take constructive methods for the salesman traveling problem as a test example, we show that the constructive algorithm obtained by AEL outperforms simple hand-crafted and LLM-generated heuristics. Compared with other domain deep learning model-based algorithms, these methods exhibit excellent scalability across different problem sizes. AEL is also very different from previous attempts that utilize LLMs as search operators in algorithms.

In artificial-intelligence-aided signal processing, existing deep learning models often exhibit a black-box structure, and their validity and comprehensibility remain elusive. The integration of topological methods, despite its relatively nascent application, serves a dual purpose of making models more interpretable as well as extracting structural information from time-dependent data for smarter learning. Here, we provide a transparent and broadly applicable methodology, TopCap, to capture the most salient topological features inherent in time series for machine learning. Rooted in high-dimensional ambient spaces, TopCap is capable of capturing features rarely detected in datasets with low intrinsic dimensionality. Applying time-delay embedding and persistent homology, we obtain descriptors which encapsulate information such as the vibration of a time series, in terms of its variability of frequency, amplitude, and average line, demonstrated with simulated data. This information is then vectorised and fed into multiple machine learning algorithms such as k-nearest neighbours and support vector machine. Notably, in classifying voiced and voiceless consonants, TopCap achieves an accuracy exceeding 96% and is geared towards designing topological convolutional layers for deep learning of speech and audio signals.