Integration of Unmanned Aerial Vehicles (UAVs) or "drones" into the civil aviation airspace is a problem of increasing interest in the aviation community, as testified by many initiatives developed worldwide. Many traditional surveillance solutions for manned aircrafts employ the Automatic Dependent System-Broadcast (ADS-B) technology, which however might present several drawbacks when used for UAVs, especially smaller ones and/or those flying at very low altitudes. We present in this paper a cloud-based surveillance solution for UAVs, which can be considered as an enhancement of a conventional ADS-B system. The proposed solution leverages inexpensive on-board transceivers for transmitting positional messages from the UAVs to the ground. A network of ADS-B gateways, based on the software-defined radio (SDR) paradigm, format the positional messages into valid ADS-B signals and rebroadcast them in the air, allowing thus to emulate a true ADS-B system and overcoming the main disadvantages of the conventional implementation. A preliminary performance analysis of the proposed approach, based on queuing theory, shows the main tradeoffs of the considered approach. Moreover, a physical-layer laboratory implementation of the proposed solution is presented, based on off-the-shelf SDR hardware, which is programmed using the open-source GNU Radio environment.
We develop Temporal Quantile Adjustment (TQA), a general method to construct efficient and valid prediction intervals (PIs) for regression on cross-sectional time series data. Such data is common in many domains, including econometrics and healthcare. A canonical example in healthcare is predicting patient outcomes using physiological time-series data, where a population of patients composes a cross-section. Reliable PI estimators in this setting must address two distinct notions of coverage: cross-sectional coverage across a cross-sectional slice, and longitudinal coverage along the temporal dimension for each time series. Recent works have explored adapting Conformal Prediction (CP) to obtain PIs in the time series context. However, none handles both notions of coverage simultaneously. CP methods typically query a pre-specified quantile from the distribution of nonconformity scores on a calibration set. TQA adjusts the quantile to query in CP at each time $t$, accounting for both cross-sectional and longitudinal coverage in a theoretically-grounded manner. The post-hoc nature of TQA facilitates its use as a general wrapper around any time series regression model. We validate TQA's performance through extensive experimentation: TQA generally obtains efficient PIs and improves longitudinal coverage while preserving cross-sectional coverage.
The number of agents can be an effective curriculum variable for controlling the difficulty of multi-agent reinforcement learning (MARL) tasks. Existing work typically uses manually defined curricula such as linear schemes. We identify two potential flaws while applying existing reward-based automatic curriculum learning methods in MARL: (1) The expected episode return used to measure task difficulty has high variance; (2) Credit assignment difficulty can be exacerbated in tasks where increasing the number of agents yields higher returns which is common in many MARL tasks. To address these issues, we propose to control the curriculum by using a TD-error based *learning progress* measure and by letting the curriculum proceed from an initial context distribution to the final task specific one. Since our approach maintains a distribution over the number of agents and measures learning progress rather than absolute performance, which often increases with the number of agents, we alleviate problem (2). Moreover, the learning progress measure naturally alleviates problem (1) by aggregating returns. In three challenging sparse-reward MARL benchmarks, our approach outperforms state-of-the-art baselines.
Representational distinctions within categories are important in all perceptual modalities and also in cognitive and motor representations. Recent pattern-information studies of brain activity have used condition-rich designs to sample the stimulus space more densely. To test whether brain response patterns discriminate among a set of stimuli (e.g. exemplars within a category) with good sensitivity, we can pool statistical evidence over all pairwise comparisons. Here we describe a wide range of statistical tests of exemplar discriminability and assess the validity (specificity) and power (sensitivity) of each test. The tests include previously used and novel, parametric and nonparametric tests, which treat subject as a random or fixed effect, and are based on different dissimilarity measures, different test statistics, and different inference procedures. We use simulated and real data to determine which tests are valid and which are most sensitive. A popular test statistic reflecting exemplar information is the exemplar discriminability index (EDI), which is defined as the average of the pattern dissimilarity estimates between different exemplars minus the average of the pattern dissimilarity estimates between repetitions of identical exemplars. The popular across-subject t test of the EDI (typically using correlation distance as the pattern dissimilarity measure) requires the assumption that the EDI is 0-mean normal under H0. Although this assumption is not strictly true, our simulations suggest that the test controls the false-positives rate at the nominal level, and is thus valid, in practice. However, test statistics based on average Mahalanobis distances or average linear-discriminant t values (both accounting for the multivariate error covariance among responses) are substantially more powerful for both random- and fixed-effects inference.
Ponzi schemes that offer absurdly high rates of return by relying on more and more people paying into the scheme have been documented since at least the mid-1800s. Ponzi schemes have shifted online in the Internet age, and some are re-branded as HYIPs or High Yield Investment Programs. This paper focuses on understanding HYIPs' continuous presence and presents various possible reasons behind their existence in today's world. A look into the countries where these schemes purport to exist, we find that 62.89% of all collected HYIPs claim to be in the United Kingdom (UK), and a further 55.56% are officially registered in the UK as a 'limited company' with a registration number provided by the UK Companies House, a UK agency that registers companies. We investigate other factors influencing these schemes, including the HYIPs' social media platforms and payment processors. The lifetime of the HYIPs helps to understand the success/failure of the investment schemes and helps indicate the schemes that could attract more investors. Using Cox proportional regression analysis, we find that having a valid UK address significantly affects the lifetime of an HYIP.
To provide rigorous uncertainty quantification for online learning models, we develop a framework for constructing uncertainty sets that provably control risk -- such as coverage of confidence intervals, false negative rate, or F1 score -- in the online setting. This extends conformal prediction to apply to a larger class of online learning problems. Our method guarantees risk control at any user-specified level even when the underlying data distribution shifts drastically, even adversarially, over time in an unknown fashion. The technique we propose is highly flexible as it can be applied with any base online learning algorithm (e.g., a deep neural network trained online), requiring minimal implementation effort and essentially zero additional computational cost. We further extend our approach to control multiple risks simultaneously, so the prediction sets we generate are valid for all given risks. To demonstrate the utility of our method, we conduct experiments on real-world tabular time-series data sets showing that the proposed method rigorously controls various natural risks. Furthermore, we show how to construct valid intervals for an online image-depth estimation problem that previous sequential calibration schemes cannot handle.
This study utilized ecologically valid social media data to identify motivations and relevant topics regarding the interaction with conversational artificial intelligence (AI) in a natural setting through investigating user conversations on Reddit, a social mediated community of practice. By applying the latent Dirichlet allocation approach, this study extracted conversation topics in six subreddit communities among users of AI–powered virtual assistants (Apple Siri, Amazon Alexa, and Google Assistant) and the corresponding smart speakers (Apple HomePod, Amazon Echo, and Google Home), and investigated changes in the conversation topics over time. Findings showed six themes of conversation topics, i.e., functional gratification, hedonic gratification, social gratification, settings, problems encountered, and connections between devices. A large volatility of the conversation topics over time was found. The results implied that members in subreddit communities share their motivations for interacting with conversational AI and collaboratively discuss the relevant issues and problem solving to learn how to practice better.
Coreference resolution -- which is a crucial task for understanding discourse and language at large -- has yet to witness widespread benefits from large language models (LLMs). Moreover, coreference resolution systems largely rely on supervised labels, which are highly expensive and difficult to annotate, thus making it ripe for prompt engineering. In this paper, we introduce a QA-based prompt-engineering method and discern \textit{generative}, pre-trained LLMs' abilities and limitations toward the task of coreference resolution. Our experiments show that GPT-2 and GPT-Neo can return valid answers, but that their capabilities to identify coreferent mentions are limited and prompt-sensitive, leading to inconsistent results.
L2 regularization for weights in neural networks is widely used as a standard training trick. However, L2 regularization for gamma, a trainable parameter of batch normalization, remains an undiscussed mystery and is applied in different ways depending on the library and practitioner. In this paper, we study whether L2 regularization for gamma is valid. To explore this issue, we consider two approaches: 1) variance control to make the residual network behave like identity mapping and 2) stable optimization through the improvement of effective learning rate. Through two analyses, we specify the desirable and undesirable gamma to apply L2 regularization and propose four guidelines for managing them. In several experiments, we observed the increase and decrease in performance caused by applying L2 regularization to gamma of four categories, which is consistent with our four guidelines. Our proposed guidelines were validated through various tasks and architectures, including variants of residual networks and transformers.
In this work we derive new analytic expressions for fixation time in Wright-Fisher model with selection. The three standard cases for fixation are considered: fixation to zero, to one or both. Second order differential equations for fixation time are obtained by a simplified approach using only the law of total probability and Taylor expansions. The obtained solutions are given by a combination of exponential integral functions with elementary functions. We then state approximate formulas involving only elementary functions valid for small selection effects. The quality of our results are explored throughout an extensive simulation study. We show that our results approximate the discrete problem very accurately even for small population size (a few hundreds) and large selection coefficients.
Detection of extremely rare variant alleles, such as tumour DNA, within a complex mixture of DNA molecules is experimentally challenging due to sequencing errors. Barcoding of target DNA molecules in library construction for next-generation sequencing provides a way to identify and bioinformatically remove polymerase induced errors. During the barcoding procedure involving $t$ consecutive PCR cycles, the DNA molecules become barcoded by unique molecular identifiers (UMI). Different library construction protocols utilise different values of $t$. The effect of a larger $t$ and imperfect PCR amplifications is poorly described. This paper proposes a branching process with growing immigration as a model describing the random outcome of $t$ cycles of PCR barcoding. Our model discriminates between five different amplification rates $r_1$, $r_2$, $r_3$, $r_4$, $r$ for different types of molecules associated with the PCR barcoding procedure. We study this model by focussing on $C_t$, the number of clusters of molecules sharing the same UMI, as well as $C_t(m)$, the number of UMI clusters of size $m$. Our main finding is a remarkable asymptotic pattern valid for moderately large $t$. It turns out that $E(C_t(m))/E(C_t)\approx 2^{-m}$ for $m=1,2,\ldots$, regardless of the underlying parameters $(r_1,r_2,r_3,r_4,r)$. The knowledge of the quantities $C_t$ and $C_t(m)$ as functions of the experimental parameters $t$ and $(r_1,r_2,r_3,r_4,r)$ will help the users to draw more adequate conclusions from the outcomes of different sequencing protocols.
Few-shot learning (FSL) has emerged as an effective learning method and shows great potential. Despite the recent creative works in tackling FSL tasks, learning valid information rapidly from just a few or even zero samples still remains a serious challenge. In this context, we extensively investigated 200+ latest papers on FSL published in the past three years, aiming to present a timely and comprehensive overview of the most recent advances in FSL along with impartial comparisons of the strengths and weaknesses of the existing works. For the sake of avoiding conceptual confusion, we first elaborate and compare a set of similar concepts including few-shot learning, transfer learning, and meta-learning. Furthermore, we propose a novel taxonomy to classify the existing work according to the level of abstraction of knowledge in accordance with the challenges of FSL. To enrich this survey, in each subsection we provide in-depth analysis and insightful discussion about recent advances on these topics. Moreover, taking computer vision as an example, we highlight the important application of FSL, covering various research hotspots. Finally, we conclude the survey with unique insights into the technology evolution trends together with potential future research opportunities in the hope of providing guidance to follow-up research.
Understanding causality has vital importance for various Natural Language Processing (NLP) applications. Beyond the labeled instances, conceptual explanations of the causality can provide deep understanding of the causal facts to facilitate the causal reasoning process. However, such explanation information still remains absent in existing causal reasoning resources. In this paper, we fill this gap by presenting a human-annotated explainable CAusal REasoning dataset (e-CARE), which contains over 21K causal reasoning questions, together with natural language formed explanations of the causal questions. Experimental results show that generating valid explanations for causal facts still remains especially challenging for the state-of-the-art models, and the explanation information can be helpful for promoting the accuracy and stability of causal reasoning models.
High-rank line-of-sight (LOS) MIMO systems have attracted considerable attention for millimeter wave and THz communications. The small wavelengths in these frequencies enable spatial multiplexing with massive data rates at long distances. Such systems are also being considered for multi-path non-LOS (NLOS) environments. In these scenarios, standard channel models based on plane waves cannot capture the curvature of each wave front necessary to model spatial multiplexing. This work presents a novel and simple multi-path wireless channel parametrization where each path is replaced by a LOS path with a reflected image source. The model fully is valid for all paths with specular planar reflections and captures the spherical nature of each wave front. Importantly, it is shown that the model uses only two additional parameters relative to the standard plane wave model. Moreover, the parameters can be easily captured in standard ray tracing. The accuracy of the approach is demonstrated on detailed ray tracing simulations at 28 GHz and 140 GHz in a dense urban area.
We consider the problem of learning personalized treatment policies that are externally valid or generalizable: they perform well in other target populations besides the experimental (or training) population from which data are sampled. We first show that welfare-maximizing policies for the experimental population are robust to shifts in the distribution of outcomes (but not characteristics) between the experimental and target populations. We then develop new methods for learning policies that are robust to shifts in outcomes and characteristics. In doing so, we highlight how treatment effect heterogeneity within the experimental population affects the generalizability of policies. Our methods may be used with experimental or observational data (where treatment is endogenous). Many of our methods can be implemented with linear programming.
Control barrier functions (CBFs) are widely used in safety-critical controllers. However, constructing a valid CBF is challenging, especially under nonlinear or non-convex constraints and for high relative degree systems. Meanwhile, finding a conservative CBF that only recovers a portion of the true safe set is usually possible. In this work, starting from a "conservative" handcrafted CBF (HCBF), we develop a method to find a CBF that recovers a reasonably larger portion of the safe set. Since the learned CBF controller is not guaranteed to be safe during training iterations, we use a model predictive controller (MPC) to ensure safety during training. Using the collected trajectory data containing safe and unsafe interactions, we train a neural network to estimate the difference between the HCBF and a CBF that recovers a closer solution to the true safe set. With our proposed approach, we can generate safe controllers that are less conservative and computationally more efficient. We validate our approach on two systems: a second-order integrator and a ball-on-beam.
Diagnostic tests for regression discontinuity design face a size-control problem. We document a massive over-rejection of the diagnostic restriction among empirical studies in the top five economics journals. At least one diagnostic test was rejected for 19 out of 59 studies, whereas less than 5% of the collected 787 tests rejected the null hypotheses. In other words, one-third of the studies rejected at least one of their diagnostic tests, whereas their underlying identifying restrictions appear plausible. Multiple testing causes this problem because the median number of tests per study was as high as 12. Therefore, we offer unified tests to overcome the size-control problem. Our procedure is based on the new joint asymptotic normality of local polynomial mean and density estimates. In simulation studies, our unified tests outperformed the Bonferroni correction. We implement the procedure as an R package rdtest with two empirical examples in its vignettes.
The machine translation (MT) task is typically formulated as that of returning a single translation for an input segment. However, in many cases, multiple different translations are valid and the appropriate translation may depend on the intended target audience, characteristics of the speaker, or even the relationship between speakers. Specific problems arise when dealing with honorifics, particularly translating from English into languages with formality markers. For example, the sentence "Are you sure?" can be translated in German as "Sind Sie sich sicher?" (formal register) or "Bist du dir sicher?" (informal). Using wrong or inconsistent tone may be perceived as inappropriate or jarring for users of certain cultures and demographics. This work addresses the problem of learning to control target language attributes, in this case formality, from a small amount of labeled contrastive data. We introduce an annotated dataset (CoCoA-MT) and an associated evaluation metric for training and evaluating formality-controlled MT models for six diverse target languages. We show that we can train formality-controlled models by fine-tuning on labeled contrastive data, achieving high accuracy (82% in-domain and 73% out-of-domain) while maintaining overall quality.
Knowing the reasoning chains from knowledge to the predicted answers can help construct an explainable question answering (QA) system. Advances on QA explanation propose to explain the answers with entailment trees composed of multiple entailment steps. While current work proposes to generate entailment trees with end-to-end generative models, the steps in the generated trees are not constrained and could be unreliable. In this paper, we propose METGEN, a Module-based Entailment Tree GENeration framework that has multiple modules and a reasoning controller. Given a question and several supporting knowledge, METGEN can iteratively generate the entailment tree by conducting single-step entailment with separate modules and selecting the reasoning flow with the controller. As each module is guided to perform a specific type of entailment reasoning, the steps generated by METGEN are more reliable and valid. Experiment results on the standard benchmark show that METGEN can outperform previous state-of-the-art models with only 9% of the parameters.
The statistical analysis of univariate quantiles is a well developed research topic. However, there is a need for research in multivariate quantiles. We construct bivariate (conditional) quantiles using the level curves of vine copula based bivariate regression model. Vine copulas are graph theoretical models identified by a sequence of linked trees, which allow for separate modelling of marginal distributions and the dependence structure. We introduce a novel graph structure model (given by a tree sequence) specifically designed for a symmetric treatment of two responses in a predictive regression setting. We establish computational tractability of the model and a straight forward way of obtaining different conditional distributions. Using vine copulas the typical shortfalls of regression, as the need for transformations or interactions of predictors, collinearity or quantile crossings are avoided. We illustrate the copula based bivariate level curves for different copula distributions and show how they can be adjusted to form valid quantile curves. We apply our approach to weather measurements from Seoul, Korea. This data example emphasizes the benefits of the joint bivariate response modelling in contrast to two separate univariate regressions or by assuming conditional independence, for bivariate response data set in the presence of conditional dependence.