With the development of blockchain technology, more and more attention has been paid to the intersection of blockchain and education, and various educational evaluation systems and E-learning systems are developed based on blockchain technology. Among them, Ethereum smart contract is favored by developers for its ``event-triggered" mechanism for building education intelligent trading systems and intelligent learning platforms. However, due to the immutability of blockchain, published smart contracts cannot be modified, so problematic contracts cannot be fixed by modifying the code in the educational blockchain. In recent years, security incidents due to smart contract vulnerabilities have caused huge property losses, so the detection of smart contract vulnerabilities in educational blockchain has become a great challenge. To solve this problem, this paper proposes a graph neural network (GNN) based vulnerability detection for smart contracts in educational blockchains. Firstly, the bytecodes are decompiled to get the opcode. Secondly, the basic blocks are divided, and the edges between the basic blocks according to the opcode execution logic are added. Then, the control flow graphs (CFG) are built. Finally, we designed a GNN-based model for vulnerability detection. The experimental results show that the proposed method is effective for the vulnerability detection of smart contracts. Compared with the traditional approaches, it can get good results with fewer layers of the GCN model, which shows that the contract bytecode and GCN model are efficient in vulnerability detection.

This paper addresses the problem of set-to-set matching, which involves matching two different sets of items based on some criteria, especially in the case of high-dimensional items like images. Although neural networks have been applied to solve this problem, most machine learning-based approaches assume that the training and test data follow the same distribution, which is not always true in real-world scenarios. To address this limitation, we introduce SHIFT15M, a dataset that can be used to evaluate set-to-set matching models when the distribution of data changes between training and testing. We conduct benchmark experiments that demonstrate the performance drop of naive methods due to distribution shift. Additionally, we provide software to handle the SHIFT15M dataset in a simple manner, with the URL for the software to be made available after publication of this manuscript. We believe proposed SHIFT15M dataset provide a valuable resource for evaluating set-to-set matching models under the distribution shift.

Bayesian Causal Forests (BCF) is a causal inference machine learning model based on a highly flexible non-parametric regression and classification tool called Bayesian Additive Regression Trees (BART). Motivated by data from the Trends in International Mathematics and Science Study (TIMSS), which includes data on student achievement in both mathematics and science, we present a multivariate extension of the BCF algorithm. With the help of simulation studies we show that our approach can accurately estimate causal effects for multiple outcomes subject to the same treatment. We also apply our model to Irish data from TIMSS 2019. Our findings reveal the positive effects of having access to a study desk at home (Mathematics ATE 95% CI: [0.20, 11.67]) while also highlighting the negative consequences of students often feeling hungry at school (Mathematics ATE 95% CI: [-11.15, -2.78] , Science ATE 95% CI: [-10.82,-1.72]) or often being absent (Mathematics ATE 95% CI: [-12.47, -1.55]).

Mathematical reasoning skills are essential for general-purpose intelligent systems to perform tasks from grocery shopping to climate modeling. Towards evaluating and improving AI systems in this domain, we propose LILA, a unified mathematical reasoning benchmark consisting of 23 diverse tasks along four dimensions: (i) mathematical abilities e.g., arithmetic, calculus (ii) language format e.g., question-answering, fill-in-the-blanks (iii) language diversity e.g., no language, simple language (iv) external knowledge e.g., commonsense, physics. We construct our benchmark by extending 20 datasets benchmark by collecting task instructions and solutions in the form of Python programs, thereby obtaining explainable solutions in addition to the correct answer. We additionally introduce two evaluation datasets to measure out-of-distribution performance and robustness to language perturbation. Finally, we introduce BHASKARA, a general-purpose mathematical reasoning model trained on LILA. Importantly, we find that multi-tasking leads to significant improvements (average relative improvement of 21.83% F1 score vs. single-task models), while the best performing model only obtains 60.40%, indicating the room for improvement in general mathematical reasoning and understanding.

In the experts problem, on each of $T$ days, an agent needs to follow the advice of one of $n$ ``experts''. After each day, the loss associated with each expert's advice is revealed. A fundamental result in learning theory says that the agent can achieve vanishing regret, i.e. their cumulative loss is within $o(T)$ of the cumulative loss of the best-in-hindsight expert. Can the agent perform well without sufficient space to remember all the experts? We extend a nascent line of research on this question in two directions: $\bullet$ We give a new algorithm against the oblivious adversary, improving over the memory-regret tradeoff obtained by [PZ23], and nearly matching the lower bound of [SWXZ22]. $\bullet$ We also consider an adaptive adversary who can observe past experts chosen by the agent. In this setting we give both a new algorithm and a novel lower bound, proving that roughly $\sqrt{n}$ memory is both necessary and sufficient for obtaining $o(T)$ regret.

Tried-and-true flapping wing robot simulation is essential in developing flapping wing mechanisms and algorithms. This paper presents a novel application-oriented flapping wing platform, highly compatible with various mechanical designs and adaptable to different robotic tasks. First, the blade element theory and the quasi-steady model are put forward to compute the flapping wing aerodynamics based on wing kinematics. Translational lift, translational drag, rotational lift, and added mass force are all considered in the computation. Then we use the proposed simulation platform to investigate the passive wing rotation and the wing-tail interaction phenomena of a particular flapping-wing robot. With the help of the simulation tool and a novel statistic based on dynamic differences from the averaged system, several behaviors display their essence by investigating the flapping wing robot dynamic characteristics. After that, the attitude tracking control problem and the positional trajectory tracking problem are both overcome by robust control techniques. Further comparison simulations reveal that the proposed control algorithms compared with other existing ones show apparent superiority. What is more, with the same control algorithm and parameters tuned in simulation, we conduct real flight experiments on a self-made flapping wing robot, and obtain similar results from the proposed simulation platform. In contrast to existing simulation tools, the proposed one is compatible with most existing flapping wing robots, and can inherently drill into each subtle behavior in corresponding applications by observing aerodynamic forces and torques on each blade element.

Gathering real-world data from the robot quickly becomes a bottleneck when constructing a robot learning system for grasping. In this work, we design a semi-supervised grasping system that, on top of a small sample of robot experience, takes advantage of images of products to be picked, which are collected without any interactions with the robot. We validate our findings both in the simulation and in the real world. In the regime of a small number of robot training samples, taking advantage of the unlabeled data allows us to achieve performance at the level of 10-fold bigger dataset size used by the baseline. The code and datasets used in the paper will be released at https://github.com/nomagiclab/grasping-student.

Restless multi-armed bandits (RMABs) extend multi-armed bandits to allow for stateful arms, where the state of each arm evolves restlessly with different transitions depending on whether that arm is pulled. Solving RMABs requires information on transition dynamics, which are often unknown upfront. To plan in RMAB settings with unknown transitions, we propose the first online learning algorithm based on the Whittle index policy, using an upper confidence bound (UCB) approach to learn transition dynamics. Specifically, we estimate confidence bounds of the transition probabilities and formulate a bilinear program to compute optimistic Whittle indices using these estimates. Our algorithm, UCWhittle, achieves sublinear $O(H \sqrt{T \log T})$ frequentist regret to solve RMABs with unknown transitions in $T$ episodes with a constant horizon $H$. Empirically, we demonstrate that UCWhittle leverages the structure of RMABs and the Whittle index policy solution to achieve better performance than existing online learning baselines across three domains, including one constructed from a real-world maternal and childcare dataset.

We study the problem of {\em online} low-rank matrix completion with $\mathsf{M}$ users, $\mathsf{N}$ items and $\mathsf{T}$ rounds. In each round, the algorithm recommends one item per user, for which it gets a (noisy) reward sampled from a low-rank user-item preference matrix. The goal is to design a method with sub-linear regret (in $\mathsf{T}$) and nearly optimal dependence on $\mathsf{M}$ and $\mathsf{N}$. The problem can be easily mapped to the standard multi-armed bandit problem where each item is an {\em independent} arm, but that leads to poor regret as the correlation between arms and users is not exploited. On the other hand, exploiting the low-rank structure of reward matrix is challenging due to non-convexity of the low-rank manifold. We first demonstrate that the low-rank structure can be exploited using a simple explore-then-commit (ETC) approach that ensures a regret of $O(\mathsf{polylog} (\mathsf{M}+\mathsf{N}) \mathsf{T}^{2/3})$. That is, roughly only $\mathsf{polylog} (\mathsf{M}+\mathsf{N})$ item recommendations are required per user to get a non-trivial solution. We then improve our result for the rank-$1$ setting which in itself is quite challenging and encapsulates some of the key issues. Here, we propose \textsc{OCTAL} (Online Collaborative filTering using iterAtive user cLustering) that guarantees nearly optimal regret of $O(\mathsf{polylog} (\mathsf{M}+\mathsf{N}) \mathsf{T}^{1/2})$. OCTAL is based on a novel technique of clustering users that allows iterative elimination of items and leads to a nearly optimal minimax rate.

Observing a human demonstrator manipulate objects provides a rich, scalable and inexpensive source of data for learning robotic policies. However, transferring skills from human videos to a robotic manipulator poses several challenges, not least a difference in action and observation spaces. In this work, we use unlabeled videos of humans solving a wide range of manipulation tasks to learn a task-agnostic reward function for robotic manipulation policies. Thanks to the diversity of this training data, the learned reward function sufficiently generalizes to image observations from a previously unseen robot embodiment and environment to provide a meaningful prior for directed exploration in reinforcement learning. We propose two methods for scoring states relative to a goal image: through direct temporal regression, and through distances in an embedding space obtained with time-contrastive learning. By conditioning the function on a goal image, we are able to reuse one model across a variety of tasks. Unlike prior work on leveraging human videos to teach robots, our method, Human Offline Learned Distances (HOLD) requires neither a priori data from the robot environment, nor a set of task-specific human demonstrations, nor a predefined notion of correspondence across morphologies, yet it is able to accelerate training of several manipulation tasks on a simulated robot arm compared to using only a sparse reward obtained from task completion.