Natural language processing (NLP) is utilized in a wide range of fields, where words in text are typically transformed into feature vectors called embeddings. BioConceptVec is a specific example of embeddings tailored for biology, trained on approximately 30 million PubMed abstracts using models such as skip-gram. Generally, word embeddings are known to solve analogy tasks through simple vector arithmetic. For instance, $\mathrm{\textit{king}} - \mathrm{\textit{man}} + \mathrm{\textit{woman}}$ predicts $\mathrm{\textit{queen}}$. In this study, we demonstrate that BioConceptVec embeddings, along with our own embeddings trained on PubMed abstracts, contain information about drug-gene relations and can predict target genes from a given drug through analogy computations. We also show that categorizing drugs and genes using biological pathways improves performance. Furthermore, we illustrate that vectors derived from known relations in the past can predict unknown future relations in datasets divided by year.

To address the peculiarities of directed and/or signed graphs, new Laplacian operators have emerged. For instance, in the case of directionality, we encounter the magnetic operator, dilation (which is underexplored), operators based on random walks, and so forth. The definition of these new operators leads to the need for new studies and concepts, and consequently, the development of new computational tools. But is this really necessary? In this work, we define the concept of effective adjacency matrices that arise from the definition of deformed Laplacian operators such as magnetic, dilation, and signal. These effective matrices allow mapping generic graphs to a family of unsigned, undirected graphs, enabling the application of the well-explored toolkit of measures, machine learning methods, and renormalization groups of undirected graphs. To explore the interplay between deformed operators and effective matrices, we show how the Hodge-Helmholtz decomposition can assist us in navigating this complexity.

Replica exchange stochastic gradient Langevin dynamics (reSGLD) is an effective sampler for non-convex learning in large-scale datasets. However, the simulation may encounter stagnation issues when the high-temperature chain delves too deeply into the distribution tails. To tackle this issue, we propose reflected reSGLD (r2SGLD): an algorithm tailored for constrained non-convex exploration by utilizing reflection steps within a bounded domain. Theoretically, we observe that reducing the diameter of the domain enhances mixing rates, exhibiting a $\textit{quadratic}$ behavior. Empirically, we test its performance through extensive experiments, including identifying dynamical systems with physical constraints, simulations of constrained multi-modal distributions, and image classification tasks. The theoretical and empirical findings highlight the crucial role of constrained exploration in improving the simulation efficiency.

In various applications, the optimal policy in a strategic decision-making problem depends both on the environmental configuration and exogenous events. For these settings, we introduce Bilevel Optimization with Contextual Markov Decision Processes (BO-CMDP), a stochastic bilevel decision-making model, where the lower level consists of solving a contextual Markov Decision Process (CMDP). BO-CMDP can be viewed as a Stackelberg Game where the leader and a random context beyond the leader's control together decide the setup of (many) MDPs that (potentially multiple) followers best respond to. This framework extends beyond traditional bilevel optimization and finds relevance in diverse fields such as model design for MDPs, tax design, reward shaping and dynamic mechanism design. We propose a stochastic Hyper Policy Gradient Descent (HPGD) algorithm to solve BO-CMDP, and demonstrate its convergence. Notably, HPGD only utilizes observations of the followers' trajectories. Therefore, it allows followers to use any training procedure and the leader to be agnostic of the specific algorithm used, which aligns with various real-world scenarios. We further consider the setting when the leader can influence the training of followers and propose an accelerated algorithm. We empirically demonstrate the performance of our algorithm.

Actor-critic (AC) is a powerful method for learning an optimal policy in reinforcement learning, where the critic uses algorithms, e.g., temporal difference (TD) learning with function approximation, to evaluate the current policy and the actor updates the policy along an approximate gradient direction using information from the critic. This paper provides the \textit{tightest} non-asymptotic convergence bounds for both the AC and natural AC (NAC) algorithms. Specifically, existing studies show that AC converges to an $\epsilon+\varepsilon_{\text{critic}}$ neighborhood of stationary points with the best known sample complexity of $\mathcal{O}(\epsilon^{-2})$ (up to a log factor), and NAC converges to an $\epsilon+\varepsilon_{\text{critic}}+\sqrt{\varepsilon_{\text{actor}}}$ neighborhood of the global optimum with the best known sample complexity of $\mathcal{O}(\epsilon^{-3})$, where $\varepsilon_{\text{critic}}$ is the approximation error of the critic and $\varepsilon_{\text{actor}}$ is the approximation error induced by the insufficient expressive power of the parameterized policy class. This paper analyzes the convergence of both AC and NAC algorithms with compatible function approximation. Our analysis eliminates the term $\varepsilon_{\text{critic}}$ from the error bounds while still achieving the best known sample complexities. Moreover, we focus on the challenging single-loop setting with a single Markovian sample trajectory. Our major technical novelty lies in analyzing the stochastic bias due to policy-dependent and time-varying compatible function approximation in the critic, and handling the non-ergodicity of the MDP due to the single Markovian sample trajectory. Numerical results are also provided in the appendix.

Fox News Flash top headlines are here. Check out what's clicking on Foxnews.com. Fire broke out at an oil refinery in northwestern Russia on Sunday, resulting in deaths and injuries, local officials said. The regional governor said the fire was not caused by a Ukrainian drone strike and investigators opened a criminal case on suspicion of negligence. The fire near the city of Ukhta in Russia's northwestern Komi Republic left at least three people injured, Komi's emergencies ministry said.

Former Director of National Intelligence John Ratcliffe weighs in fallout following the conviction of Donald Trump and his expectations for the upcoming sentencing. The Justice Department cannot release audio from President Biden's interview with Special Counsel Robert Hur due to the threat of potential deepfakes, the DOJ argued in a Friday court filing. The filing came as part of a legal challenge against Biden's efforts to exercise executive privilege over the recording to keep it from the public. The DOJ acknowledged in its Friday filing that there is already enough public audio available to create AI deepfakes of both Biden and Hur, but it said releasing the true recording would make it more difficult to disprove any false versions. "The passage of time and advancements in audio, artificial intelligence, and'deep fake' technologies only amplify concerns about malicious manipulation of audio files. If the audio recording is released here, it is easy to foresee that it could be improperly altered, and that the altered file could be passed off as an authentic recording and widely distributed," the department wrote.

Actor Toby Jones, who played the lead role in the TV series about the Post Office scandal, has said he "played a hero" in the drama. He starred as Alan Bates in the ITV show Mr Bates vs the Post Office, which tells the story of hundreds of sub-postmasters who were wrongly accused of theft, fraud and false accounting. Many of the reported losses in branch accounts which led to prosecutions were actually caused by faults in Horizon - the Post Office's computer system. Jones, speaking at the Hay Festival, said Bates - a former sub-postmaster himself, who led the victims' campaign to get justice - "doesn't want any honours until he's finished the job". "I get to play a hero. Really, someone who I think of as a hero. Someone in the culture who just doesn't seem to be subject to the same forces that we all are," the 57-year-old said.

The advent of Intelligent Reflecting Surfaces (IRS) and Unmanned Aerial Vehicles (UAVs) is setting a new benchmark in the field of wireless communications. IRS, with their groundbreaking ability to manipulate electromagnetic waves, have opened avenues for substantial enhancements in signal quality, network efficiency, and spectral usage. These surfaces dynamically reconfigure the propagation environment, leading to optimized signal paths and reduced interference. Concurrently, UAVs have emerged as dynamic, versatile elements within communication networks, offering high mobility and the ability to access and enhance coverage in areas where traditional, fixed infrastructure falls short. This paper presents a comprehensive survey on the synergistic integration of IRS and UAVs in wireless networks, highlighting how this innovative combination substantially boosts network performance, particularly in terms of security, energy efficiency, and reliability. The versatility of UAVs, combined with the signal-manipulating prowess of IRS, creates a potent solution for overcoming the limitations of conventional communication setups, especially in challenging and underserved environments. Furthermore, the survey delves into the cutting-edge realm of Machine Learning (ML), exploring its role in the strategic deployment and operational optimization of UAVs equipped with IRS. The paper also underscores the latest research and practical advancements in this field, providing insights into real-world applications and experimental setups. It concludes by discussing the future prospects and potential directions for this emerging technology, positioning the IRS-UAV integration as a transformative force in the landscape of next-generation wireless

Current language models decode text token by token according to probabilistic distribution, and determining the appropriate candidates for the next token is crucial to ensure generation quality. This study introduces adaptive decoding, a mechanism that dynamically empowers language models to ascertain a sensible candidate set during generation. Specifically, we introduce an entropy-based metric called confidence and conceptualize determining the optimal candidate set as a confidence-increasing process. The rationality of including a token in the candidate set is assessed by leveraging the increment of confidence. Experimental results reveal that our method balances diversity and coherence well. The human evaluation shows that our method can generate human-preferred text. Additionally, our method can potentially improve the reasoning ability of language models.