Online task scheduling serves an integral role for task-intensive applications in cloud computing and crowdsourcing. Optimal scheduling can enhance system performance, typically measured by the reward-to-cost ratio, under some task arrival distribution. On one hand, both reward and cost are dependent on task context (e.g., evaluation metric) and remain black-box in practice. These render reward and cost hard to model thus unknown before decision making. On the other hand, task arrival behaviors remain sensitive to factors like unpredictable system fluctuation whereby a prior estimation or the conventional assumption of arrival distribution (e.g., Poisson) may fail. This implies another practical yet often neglected challenge, i.e., uncertain task arrival distribution. Towards effective scheduling under a stationary environment with various uncertainties, we propose a double-optimistic learning based Robbins-Monro (DOL-RM) algorithm. Specifically, DOL-RM integrates a learning module that incorporates optimistic estimation for reward-to-cost ratio and a decision module that utilizes the Robbins-Monro method to implicitly learn task arrival distribution while making scheduling decisions. Theoretically, DOL-RM achieves convergence gap and no regret learning with a sub-linear regret of $O(T^{3/4})$, which is the first result for online task scheduling under uncertain task arrival distribution and unknown reward and cost. Our numerical results in a synthetic experiment and a real-world application demonstrate the effectiveness of DOL-RM in achieving the best cumulative reward-to-cost ratio compared with other state-of-the-art baselines.

Both text and video data are abundant on the internet and support large-scale self-supervised learning through next token or frame prediction. However, they have not been equally leveraged: language models have had significant real-world impact, whereas video generation has remained largely limited to media entertainment. Yet video data captures important information about the physical world that is difficult to express in language. To address this gap, we discuss an under-appreciated opportunity to extend video generation to solve tasks in the real world. We observe how, akin to language, video can serve as a unified interface that can absorb internet knowledge and represent diverse tasks. Moreover, we demonstrate how, like language models, video generation can serve as planners, agents, compute engines, and environment simulators through techniques such as in-context learning, planning and reinforcement learning. We identify major impact opportunities in domains such as robotics, self-driving, and science, supported by recent work that demonstrates how such advanced capabilities in video generation are plausibly within reach. Lastly, we identify key challenges in video generation that mitigate progress. Addressing these challenges will enable video generation models to demonstrate unique value alongside language models in a wider array of AI applications.

This work contains the mathematical exploration of a few prototypical games in which central concepts from statistics and probability theory naturally emerge. The first two kinds of games are termed Fisher and Bayesian games, which are connected to Frequentist and Bayesian statistics, respectively. Later, a more general type of game is introduced, termed Statistical game, in which a further parameter, the players' relative risk aversion, can be set. In this work, we show that Fisher and Bayesian games can be viewed as limiting cases of Statistical games. Therefore, Statistical games can be viewed as a unified framework, incorporating both Frequentist and Bayesian statistics. Furthermore, a philosophical framework is (re-)presented -- often referred to as minimax regret criterion -- as a general approach to decision making. The main motivation for this work was to embed Bayesian statistics into a broader decision-making framework, where, based on collected data, actions with consequences have to be made, which can be translated to utilities (or rewards/losses) of the decision-maker. The work starts with the simplest possible toy model, related to hypothesis testing and statistical inference. This choice has two main benefits: i.) it allows us to determine (conjecture) the behaviour of the equilibrium strategies in various limiting cases ii.) this way, we can introduce Statistical games without requiring additional stochastic parameters. The work contains game theoretical methods related to two-player, non-cooperative games to determine and prove equilibrium strategies of Fisher, Bayesian and Statistical games. It also relies on analytical tools for derivations concerning various limiting cases.

The 38th AAAI Conference on Artificial Intelligence (AAAI-24) kicked off on Tuesday 20 February. The in-person event is being held in the Vancouver Convention Centre. We take a look at what the participants have been getting up to over the past few days. We are excited to welcome everyone to #AAAI24! pic.twitter.com/UN8RUzrq1C We will present a tutorial on "Knowledge Editing for Large Language Models" at #AAAI2024 from 2-6PM PST on Feb 20.

Steering the behavior of a strong model pre-trained on internet-scale data can be difficult due to the scarcity of competent supervisors. Recent studies reveal that, despite supervisory noises, a strong student model may surpass its weak teacher when fine-tuned on specific objectives. Yet, the effectiveness of such weak-to-strong generalization remains limited, especially in the presence of large capability gaps. In this paper, we propose to address this challenge by harnessing a diverse set of specialized teachers, instead of a single generalist one, that collectively supervises the strong student. Our approach resembles the classical hierarchical mixture of experts, with two components tailored for co-supervision: (i) we progressively alternate student training and teacher assignment, leveraging the growth of the strong student to identify plausible supervisions; (ii) we conservatively enforce teacher-student and local-global consistency, leveraging their dependencies to reject potential annotation noises. We validate the proposed method through visual recognition tasks on the OpenAI weak-to-strong benchmark and additional multi-domain datasets. Our code is available at \url{https://github.com/yuejiangliu/csl}.

You've probably heard of ChatGPT by now, and heard about some of the impacts that artificial intelligence is having on the working world. AI is shaping the future, but humans still shape AI, and if you want to know how, then The Ultimate AI ChatGPT & Python Programming Bundle is for you. This bundle includes 14 courses from some of the web's top instructors, including John Elder (4.4/5-star instructor rating), Hugo Ferro (4.6/5-star rating), and Dr. Chris Mall (4.4/5-star rating). Through these courses, you'll learn how to use Python (and its various libraries) for machine learning, data science, and more. You'll also get a crash course in how to build your own chatbots that use ChatGPT to answer queries.

In the domain of Natural Language Processing (NLP), Large Language Models (LLMs) have demonstrated promise in text-generation tasks. However, their educational applications, particularly for domain-specific queries, remain underexplored. This study investigates LLMs' capabilities in educational scenarios, focusing on concept graph recovery and question-answering (QA). We assess LLMs' zero-shot performance in creating domain-specific concept graphs and introduce TutorQA, a new expert-verified NLP-focused benchmark for scientific graph reasoning and QA. TutorQA consists of five tasks with 500 QA pairs. To tackle TutorQA queries, we present CGLLM, a pipeline integrating concept graphs with LLMs for answering diverse questions. Our results indicate that LLMs' zero-shot concept graph recovery is competitive with supervised methods, showing an average 3% F1 score improvement. In TutorQA tasks, LLMs achieve up to 26% F1 score enhancement. Moreover, human evaluation and analysis show that CGLLM generates answers with more fine-grained concepts.

In online learning platforms, particularly in rapidly growing computer programming courses, addressing the thousands of students' learning queries requires considerable human cost. The creation of intelligent assistant large language models (LLMs) tailored for programming education necessitates distinct data support. However, in real application scenarios, the data resources for training such LLMs are relatively scarce. Therefore, to address the data scarcity in intelligent educational systems for programming, this paper proposes a new Chinese question-and-answer dataset for Python learners. To ensure the authenticity and reliability of the sources of the questions, we collected questions from actual student questions and categorized them according to various dimensions such as the type of questions and the type of learners. This annotation principle is designed to enhance the effectiveness and quality of online programming education, providing a solid data foundation for developing the programming teaching assists (TA). Furthermore, we conducted comprehensive evaluations of various LLMs proficient in processing and generating Chinese content, highlighting the potential limitations of general LLMs as intelligent teaching assistants in computer programming courses.

Laplacian based nonparametric regression is a widely used approach in machine learning that leverages the Laplacian Eigenmaps algorithm to perform regression tasks without relying on explicit parametric models. The nonparametric nature of the approach makes it flexible and adaptable to data generating process without imposing strict assumptions about the functional form of the relationship between the response and the covariates. Existing theoretical studies of this approach are restricted to establishing minimax rates of convergence and adaptivity properties when the true regression function lies in Sobolev spaces; see Section 1.1 for details. Such spaces are inherently smooth in nature and exclude important function classes in nonparametric statistics, such as piecewise constant or polynomial functions, bump functions and other such nonsmooth function classes. In this work, using the framework of fractional Laplacians, we propose a novel approach called Principal Component Regression using Fractional Laplacian Eigenmaps (PCR-FLE) for nonsmooth and nonparametric regression. The PCR-FLE algorithm generalizes the PCR-LE algorithm by Green et al. (2023) and the PCR-WLE algorithm by Shi et al. (2024), and is designed to naturally handle the case when the true regression function lies in an L

After being struck down by Covid, our Managing Editor Lucy Smith has been unable to make the journey to AAAI 2024 to deliver the AIhub training session on science communication for AI researchers. However, Professor Michael Wooldridge has very kindly stepped in, and will present on "How to speak to the public about AI". In this talk, Michael will elaborate on 14 lessons that he has learnt during his time communicating about AI. His vast experience in speaking about AI has ranged from media interviews, to advising the UK Government, to giving the prestigious Royal Institute Christmas Lectures. The talk will take place today (Wednesday 21 February) from 14:00 – 15:00 in room 113, Vancouver Convention Centre.