Constructing a differentially private (DP) estimator requires deriving the maximum influence of an observation, which can be difficult in the absence of exogenous bounds on the input data or the estimator, especially in high dimensional settings. This paper shows that standard notions of statistical depth, i.e., halfspace depth and regression depth, are particularly advantageous in this regard, both in the sense that the maximum influence of a single observation is easy to analyze and that this value is typically low. This is used to motivate new approximate DP location and regression estimators using the maximizers of these two notions of statistical depth. A more computationally efficient variant of the approximate DP regression estimator is also provided. Also, to avoid requiring that users specify a priori bounds on the estimates and/or the observations, variants of these DP mechanisms are described that satisfy random differential privacy (RDP), which is a relaxation of differential privacy provided by Hall, Wasserman, and Rinaldo (2013). We also provide simulations of the two DP regression methods proposed here. The proposed estimators appear to perform favorably relative to the existing DP regression methods we consider in these simulations when either the sample size is at least 100-200 or the privacy-loss budget is sufficiently high.

The approval success rate of drug candidates is very low with the majority of failure due to safety and efficacy. Increasingly available high dimensional information on targets, drug molecules and indications provides an opportunity for ML methods to integrate multiple data modalities and better predict clinically promising drug targets. Notably, drug targets with human genetics evidence are shown to have better odds to succeed. However, a recent tensor factorization-based approach found that additional information on targets and indications might not necessarily improve the predictive accuracy. Here we revisit this approach by integrating different types of human genetics evidence collated from publicly available sources to support each target-indication pair. We use Bayesian tensor factorization to show that models incorporating all available human genetics evidence (rare disease, gene burden, common disease) modestly improves the clinical outcome prediction over models using single line of genetics evidence. We provide additional insight into the relative predictive power of different types of human genetics evidence for predicting the success of clinical outcomes.

Humans have remarkable capacity to reason abductively and hypothesize about what lies beyond the literal content of an image. By identifying concrete visual clues scattered throughout a scene, we almost can't help but draw probable inferences beyond the literal scene based on our everyday experience and knowledge about the world. For example, if we see a "20 mph" sign alongside a road, we might assume the street sits in a residential area (rather than on a highway), even if no houses are pictured. Can machines perform similar visual reasoning? We present Sherlock, an annotated corpus of 103K images for testing machine capacity for abductive reasoning beyond literal image contents. We adopt a free-viewing paradigm: participants first observe and identify salient clues within images (e.g., objects, actions) and then provide a plausible inference about the scene, given the clue. In total, we collect 363K (clue, inference) pairs, which form a first-of-its-kind abductive visual reasoning dataset. Using our corpus, we test three complementary axes of abductive reasoning. We evaluate the capacity of models to: i) retrieve relevant inferences from a large candidate corpus; ii) localize evidence for inferences via bounding boxes, and iii) compare plausible inferences to match human judgments on a newly-collected diagnostic corpus of 19K Likert-scale judgments. While we find that fine-tuning CLIP-RN50x64 with a multitask objective outperforms strong baselines, significant headroom exists between model performance and human agreement. Data, models, and leaderboard available at http://visualabduction.com/

This is due to the rapid expansion of computing (Cutri et al., 2013), had many technical challenges and resources and sensor technology in the last four required intensive astronomy expertise, experience, and labor decades that has driven equally rapid expansions in the to overcome (Eisenhardt et al., 2012, for example). A quantity of data to analyze. Astronomy, in particular, necessary first step in that process, though, is to classify has seen a proliferation of large scale imaging and spectroscopic the sources so that we can prioritize which sources might surveys that have billions of sources in them-- be interesting, and which are examples of already known surveys like: the Sloan Digital Sky Survey (SDSS, York sources. Because these sources are rare it is usually easier et al., 2000), the 2-Micron All Sky Survey (2MASS, Skrutskie to use a supervised machine learning algorithm, one that et al., 2006), the Wide-field Infrared Survey Explorer is tuned using sources with known classifications, than it (WISE, Wright et al., 2010), the Gaia satellite's survey is to use an unsupervised one. The reason should be obvious: (Gaia Collaboration et al., 2016), the Panoramic Survey subgroups of the common known source types are Telescope and Rapid Response System (Pan-STARRS) likely to outnumber the rare new ones, meaning a naive surveys (Chambers et al., 2016), the Dark Energy Spectroscopic unsupervised machine learning algorithm could need a lot Instrument (DESI) surveys (Dey et al., 2019), the of complexity before it actually finds the rare class. UKIRT Infrared Deep Sky Surveys (UKIDSS, Lawrence et al., 2007), and the Galaxy Evolution Explorer (GALEX) Supervised learning also has drawbacks when used for surveys (Martin et al., 2005).

Codex, a large language model (LLM) trained on a variety of codebases, exceeds the previous state of the art in its capacity to synthesize and generate code. Although Codex provides a plethora of benefits, models that may generate code on such scale have significant limitations, alignment problems, the potential to be misused, and the possibility to increase the rate of progress in technical fields that may themselves have destabilizing impacts or have misuse potential. Yet such safety impacts are not yet known or remain to be explored. In this paper, we outline a hazard analysis framework constructed at OpenAI to uncover hazards or safety risks that the deployment of models like Codex may impose technically, socially, politically, and economically. The analysis is informed by a novel evaluation framework that determines the capacity of advanced code generation techniques against the complexity and expressivity of specification prompts, and their capability to understand and execute them relative to human ability.

Deep learning-based super-resolution (SR) has gained tremendous popularity in recent years because of its high image quality performance and wide application scenarios. However, prior methods typically suffer from large amounts of computations and huge power consumption, causing difficulties for real-time inference, especially on resource-limited platforms such as mobile devices. To mitigate this, we propose a compiler-aware SR neural architecture search (NAS) framework that conducts depth search and per-layer width search with adaptive SR blocks. The inference speed is directly taken into the optimization along with the SR loss to derive SR models with high image quality while satisfying the real-time inference requirement. Instead of measuring the speed on mobile devices at each iteration during the search process, a speed model incorporated with compiler optimizations is leveraged to predict the inference latency of the SR block with various width configurations for faster convergence. With the proposed framework, we achieve real-time SR inference for implementing 720p resolution with competitive SR performance (in terms of PSNR and SSIM) on GPU/DSP of mobile platforms (Samsung Galaxy S21).

Product-specific guidances (PSGs) recommended by the United States Food and Drug Administration (FDA) are instrumental to promote and guide generic drug product development. To assess a PSG, the FDA assessor needs to take extensive time and effort to manually retrieve supportive drug information of absorption, distribution, metabolism, and excretion (ADME) from the reference listed drug labeling. In this work, we leveraged the state-of-the-art pre-trained language models to automatically label the ADME paragraphs in the pharmacokinetics section from the FDA-approved drug labeling to facilitate PSG assessment. We applied a transfer learning approach by fine-tuning the pre-trained Bidirectional Encoder Representations from Transformers (BERT) model to develop a novel application of ADME semantic labeling, which can automatically retrieve ADME paragraphs from drug labeling instead of manual work. We demonstrated that fine-tuning the pre-trained BERT model can outperform the conventional machine learning techniques, achieving up to 11.6% absolute F1 improvement. To our knowledge, we were the first to successfully apply BERT to solve the ADME semantic labeling task. We further assessed the relative contribution of pre-training and fine-tuning to the overall performance of the BERT model in the ADME semantic labeling task using a series of analysis methods such as attention similarity and layer-based ablations. Our analysis revealed that the information learned via fine-tuning is focused on task-specific knowledge in the top layers of the BERT, whereas the benefit from the pre-trained BERT model is from the bottom layers.

"Russian forces can bring their guns to bear on the enemy within only three to five minutes of an Orlan-10 drone spotting a target," says Dr Watling. Without them, an attack could take 20 to 30 minutes to carry out, he says.

People have a natural fear of technology putting them out of work. The word sabotage allegedly comes from French protesters who threw their wooden clogs -- sabots -- into machines to stop them working. And much of the second half of the 20th century was characterised by labour disputes in relation to the introduction of new technologies in manufacturing industries. Many workers in the services and creative industries believed they were immune to such threats, but AI has changed all that. Robot process automation (RPA) and other AI-powered activities are replacing human activities in a whole range of areas, from call centres to accountancy practices; and, as the technology gets smarter, the number of roles that can be replaced increases.

Join gaming executives to discuss emerging parts of the industry this October at GamesBeat Summit Next. The world of technology is rapidly shifting from flat media viewed in the third person to immersive media experienced in the first person. Recently dubbed "the metaverse," this major transition in mainstream computing has ignited a new wave of excitement over the core technologies of virtual and augmented reality. But there is a third technology area known as telepresence that is often overlooked but will become an important part of the metaverse. While virtual reality brings users into simulated worlds, telepresence (also called telerobotics) uses remote robots to bring users to distant places, giving them the ability to look around and perform complex tasks.