Lipschitz constrained networks have gathered considerable attention in the deep learning community, with usages ranging from Wasserstein distance estimation to the training of certifiably robust classifiers. However they remain commonly considered as less accurate, and their properties in learning are still not fully understood. In this paper we clarify the matter: when it comes to classification 1-Lipschitz neural networks enjoy several advantages over their unconstrained counterpart. First, we show that these networks are as accurate as classical ones, and can fit arbitrarily difficult boundaries. Then, relying on a robustness metric that reflects operational needs we characterize the most robust classifier: the WGAN discriminator. Next, we show that 1-Lipschitz neural networks generalize well under milder assumptions. Finally, we show that hyper-parameters of the loss are crucial for controlling the accuracy-robustness trade-off. We conclude that they exhibit appealing properties to pave the way toward provably accurate, and provably robust neural networks.
A total of 605 eligible respondents took part in this survey (population size 1630046161 and required sample size 591) with an age range of 18 to 100. A large proportion of the respondents are aged less than 50 (82%) and male (62.15%). The majority of the respondents live in urban areas (60.83%). A total of 61.16% (370/605) of the respondents were willing to accept/take the COVID-19 vaccine. Among the accepted group, only 35.14% showed the willingness to take the COVID-19 vaccine immediately, while 64.86% would delay the vaccination until they are confirmed about the vaccine s efficacy and safety or COVID-19 becomes deadlier in Bangladesh. The regression results showed age, gender, location (urban/rural), level of education, income, perceived risk of being infected with COVID-19 in the future, perceived severity of infection, having previous vaccination experience after age 18, having higher knowledge about COVID-19 and vaccination were significantly associated with the acceptance of COVID-19 vaccines. The research reported a high prevalence of COVID-19 vaccine refusal and hesitancy in Bangladesh.
We explore creating automated, personalized feedback in an intelligent tutoring system (ITS). Our goal is to pinpoint correct and incorrect concepts in student answers in order to achieve better student learning gains. Although automatic methods for providing personalized feedback exist, they do not explicitly inform students about which concepts in their answers are correct or incorrect. Our approach involves decomposing students answers using neural discourse segmentation and classification techniques. This decomposition yields a relational graph over all discourse units covered by the reference solutions and student answers. We use this inferred relational graph structure and a neural classifier to match student answers with reference solutions and generate personalized feedback. Although the process is completely automated and data-driven, the personalized feedback generated is highly contextual, domain-aware and effectively targets each student's misconceptions and knowledge gaps. We test our method in a dialogue-based ITS and demonstrate that our approach results in high-quality feedback and significantly improved student learning gains.
Understanding students' misconceptions is important for effective teaching and assessment. However, discovering such misconceptions manually can be time-consuming and laborious. Automated misconception discovery can address these challenges by highlighting patterns in student data, which domain experts can then inspect to identify misconceptions. In this work, we present a novel method for the semi-automated discovery of problem-specific misconceptions from students' program code in computing courses, using a state-of-the-art code classification model. We trained the model on a block-based programming dataset and used the learned embedding to cluster incorrect student submissions. We found these clusters correspond to specific misconceptions about the problem and would not have been easily discovered with existing approaches. We also discuss potential applications of our approach and how these misconceptions inform domain-specific insights into students' learning processes.
Centralized Training for Decentralized Execution, where agents are trained offline using centralized information but execute in a decentralized manner online, has gained popularity in the multi-agent reinforcement learning community. In particular, actor-critic methods with a centralized critic and decentralized actors are a common instance of this idea. However, the implications of using a centralized critic in this context are not fully discussed and understood even though it is the standard choice of many algorithms. We therefore formally analyze centralized and decentralized critic approaches, providing a deeper understanding of the implications of critic choice. Because our theory makes unrealistic assumptions, we also empirically compare the centralized and decentralized critic methods over a wide set of environments to validate our theories and to provide practical advice. We show that there exist misconceptions regarding centralized critics in the current literature and show that the centralized critic design is not strictly beneficial, but rather both centralized and decentralized critics have different pros and cons that should be taken into account by algorithm designers.
The literature in modern machine learning has only negative results for learning to communicate between competitive agents using standard RL. We introduce a modified sender-receiver game to study the spectrum of partially-competitive scenarios and show communication can indeed emerge in a competitive setting. We empirically demonstrate three key takeaways for future research. First, we show that communication is proportional to cooperation, and it can occur for partially competitive scenarios using standard learning algorithms. Second, we highlight the difference between communication and manipulation and extend previous metrics of communication to the competitive case. Third, we investigate the negotiation game where previous work failed to learn communication between independent agents (Cao et al., 2018). We show that, in this setting, both agents must benefit from communication for it to emerge; and, with a slight modification to the game, we demonstrate successful communication between competitive agents. We hope this work overturns misconceptions and inspires more research in competitive emergent communication.
Deep neural networks have triggered a revolution in artificial intelligence, having been applied with great results in medical imaging, semi-autonomous vehicles, ecommerce, genetics research, speech recognition, particle physics, experimental art, economic forecasting, environmental science, industrial manufacturing, and a wide variety of applications in nearly every field. This sudden success, though, may have intoxicated the research community and blinded them to the potential pitfalls of assigning deep learning a higher status than warranted. Also, research directed at alleviating the weaknesses of deep learning may seem less attractive to scientists and engineers, who focus on the low-hanging fruit of finding more and more applications for deep learning models, thus letting short-term benefits hamper long-term scientific progress. Gary Marcus wrote a paper entitled Deep Learning: A Critical Appraisal, and here we discuss Marcus' core ideas, as well as attempt a general assessment of the subject. This study examines some of the limitations of deep neural networks, with the intention of pointing towards potential paths for future research, and of clearing up some metaphysical misconceptions, held by numerous researchers, that may misdirect them.
Research on quantum technology spans multiple disciplines: physics, computer science, engineering, and mathematics. The objective of this manuscript is to provide an accessible introduction to this emerging field for economists that is centered around quantum computing and quantum money. We proceed in three steps. First, we discuss basic concepts in quantum computing and quantum communication, assuming knowledge of linear algebra and statistics, but not of computer science or physics. This covers fundamental topics, such as qubits, superposition, entanglement, quantum circuits, oracles, and the no-cloning theorem. Second, we provide an overview of quantum money, an early invention of the quantum communication literature that has recently been partially implemented in an experimental setting. One form of quantum money offers the privacy and anonymity of physical cash, the option to transact without the involvement of a third party, and the efficiency and convenience of a debit card payment. Such features cannot be achieved in combination with any other form of money. Finally, we review all existing quantum speedups that have been identified for algorithms used to solve and estimate economic models. This includes function approximation, linear systems analysis, Monte Carlo simulation, matrix inversion, principal component analysis, linear regression, interpolation, numerical differentiation, and true random number generation. We also discuss the difficulty of achieving quantum speedups and comment on common misconceptions about what is achievable with quantum computing.
Driven by recent successes in two-player, zero-sum game solving and playing, artificial intelligence work on games has increasingly focused on algorithms that produce equilibrium-based strategies. However, this approach has been less effective at producing competent players in general-sum games or those with more than two players than in two-player, zero-sum games. An appealing alternative is to consider adaptive algorithms that ensure strong performance in hindsight relative to what could have been achieved with modified behavior. This approach also leads to a game-theoretic analysis, but in the correlated play that arises from joint learning dynamics rather than factored agent behavior at equilibrium. We develop and advocate for this hindsight rationality framing of learning in general sequential decision-making settings. To this end, we re-examine mediated equilibrium and deviation types in extensive-form games, thereby gaining a more complete understanding and resolving past misconceptions. We present a set of examples illustrating the distinct strengths and weaknesses of each type of equilibrium in the literature, and prove that no tractable concept subsumes all others. This line of inquiry culminates in the definition of the deviation and equilibrium classes that correspond to algorithms in the counterfactual regret minimization (CFR) family, relating them to all others in the literature. Examining CFR in greater detail further leads to a new recursive definition of rationality in correlated play that extends sequential rationality in a way that naturally applies to hindsight evaluation.
The various types of retinal neurons are each positioned at their respective depths within the retina where they are believed to be assembled as orderly mosaics, in which like-type neurons minimize proximity to one another. Two common statistical analyses for assessing the spatial properties of retinal mosaics include the nearest neighbor analysis, from which an index of their "regularity" is commonly calculated, and the density recovery profile derived from auto-correlation analysis, revealing the presence of an exclusion zone indicative of anti-clustering. While each of the spatial statistics derived from these analyses, the regularity index and the effective radius, can be useful in characterizing such properties of orderly retinal mosaics, they are rarely sufficient for conveying the natural variation in the self-spacing behavior of different types of retinal neurons and the extent to which that behavior generates uniform intercellular spacing across the mosaic. We consider the strengths and limitations of different spatial statistical analyses for assessing the patterning in retinal mosaics, highlighting a number of misconceptions and their frequent misuse. Rather than being diagnostic criteria for determining simply whether a population is "regular", they should be treated as descriptive statistics that convey variation in the factors that influence neuronal positioning. We subsequently apply multiple spatial statistics to the analysis of eight different mosaics in the mouse retina, demonstrating conspicuous variability in the degree of patterning present, from essentially random to notably regular. This variability in patterning has both a developmental as well as a functional significance, reflecting the rules governing the positioning of different types of neurons as the architecture of the retina is assembled (abstract truncated).
Preliminary detection of mild depression could immensely help in effective treatment of the common mental health disorder. Due to the lack of proper awareness and the ample mix of stigmas and misconceptions present within the society, mental health status estimation has become a truly difficult task. Due to the immense variations in character level traits from person to person, traditional deep learning methods fail to generalize in a real world setting. In our study we aim to create a human allied AI workflow which could efficiently adapt to specific users and effectively perform in real world scenarios. We propose a Hybrid deep learning approach that combines the essence of one shot learning, classical supervised deep learning methods and human allied interactions for adaptation. In order to capture maximum information and make efficient diagnosis video, audio, and text modalities are utilized. Our Hybrid Fusion model achieved a high accuracy of 96.3% on the Dataset; and attained an AUC of 0.9682 which proves its robustness in discriminating classes in complex real-world scenarios making sure that no cases of mild depression are missed during diagnosis. The proposed method is deployed in a cloud-based smartphone application for robust testing. With user-specific adaptations and state of the art methodologies, we present a state-of-the-art model with user friendly experience.
To date, there has been no formal study of the statistical cost of interpretability in machine learning. As such, the discourse around potential trade-offs is often informal and misconceptions abound. In this work, we aim to initiate a formal study of these trade-offs. A seemingly insurmountable roadblock is the lack of any agreed upon definition of interpretability. Instead, we propose a shift in perspective. Rather than attempt to define interpretability, we propose to model the \emph{act} of \emph{enforcing} interpretability. As a starting point, we focus on the setting of empirical risk minimization for binary classification, and view interpretability as a constraint placed on learning. That is, we assume we are given a subset of hypothesis that are deemed to be interpretable, possibly depending on the data distribution and other aspects of the context. We then model the act of enforcing interpretability as that of performing empirical risk minimization over the set of interpretable hypotheses. This model allows us to reason about the statistical implications of enforcing interpretability, using known results in statistical learning theory. Focusing on accuracy, we perform a case analysis, explaining why one may or may not observe a trade-off between accuracy and interpretability when the restriction to interpretable classifiers does or does not come at the cost of some excess statistical risk. We close with some worked examples and some open problems, which we hope will spur further theoretical development around the tradeoffs involved in interpretability.
The field artificial intelligence (AI) has been founded over 65 years ago. Starting with great hopes and ambitious goals the field progressed though various stages of popularity and received recently a revival in the form of deep neural networks. Some problems of AI are that so far neither 'intelligence' nor the goals of AI are formally defined causing confusion when comparing AI to other fields. In this paper, we present a perspective on the desired and current status of AI in relation to machine learning and statistics and clarify common misconceptions and myths. Our discussion is intended to uncurtain the veil of vagueness surrounding AI to see its true countenance.
Whether to give rights to artificial intelligence (AI) and robots has been a sensitive topic since the European Parliament proposed advanced robots could be granted "electronic personalities." Numerous scholars who favor or disfavor its feasibility have participated in the debate. This paper presents an experiment (N=1270) that 1) collects online users' first impressions of 11 possible rights that could be granted to autonomous electronic agents of the future and 2) examines whether debunking common misconceptions on the proposal modifies one's stance toward the issue. The results indicate that even though online users mainly disfavor AI and robot rights, they are supportive of protecting electronic agents from cruelty (i.e., favor the right against cruel treatment). Furthermore, people's perceptions became more positive when given information about rights-bearing non-human entities or myth-refuting statements. The style used to introduce AI and robot rights significantly affected how the participants perceived the proposal, similar to the way metaphors function in creating laws. For robustness, we repeated the experiment over a more representative sample of U.S. residents (N=164) and found that perceptions gathered from online users and those by the general population are similar.
Digital technologies are becoming increasingly prevalent in education, enabling personalized, high quality education resources to be accessible by students across the world. Importantly, among these resources are diagnostic questions: the answers that the students give to these questions reveal key information about the specific nature of misconceptions that the students may hold. Analyzing the massive quantities of data stemming from students' interactions with these diagnostic questions can help us more accurately understand the students' learning status and thus allow us to automate learning curriculum recommendations. In this competition, participants will focus on the students' answer records to these multiple-choice diagnostic questions, with the aim of 1) accurately predicting which answers the students provide; 2) accurately predicting which questions have high quality; and 3) determining a personalized sequence of questions for each student that best predicts the student's answers. These tasks closely mimic the goals of a real-world educational platform and are highly representative of the educational challenges faced today. We provide over 20 million examples of students' answers to mathematics questions from Eedi, a leading educational platform which thousands of students interact with daily around the globe. Participants to this competition have a chance to make a lasting, real-world impact on the quality of personalized education for millions of students across the world.
The Variational Auto-Encoder (VAE) is a simple, efficient, and popular deep maximum likelihood model. Though usage of VAEs is widespread, the derivation of the VAE is not as widely understood. In this tutorial, we will provide an overview of the VAE and a tour through various derivations and interpretations of the VAE objective. From a probabilistic standpoint, we will examine the VAE through the lens of Bayes' Rule, importance sampling, and the change-of-variables formula. From an information theoretic standpoint, we will examine the VAE through the lens of lossless compression and transmission through a noisy channel. We will then identify two common misconceptions over the VAE formulation and their practical consequences. Finally, we will visualize the capabilities and limitations of VAEs using a code example (with an accompanying Jupyter notebook) on toy 2D data.
Fine-tuning pre-trained transformer-based language models such as BERT has become a common practice dominating leaderboards across various NLP benchmarks. Despite the strong empirical performance of fine-tuned models, fine-tuning is an unstable process: training the same model with multiple random seeds can result in a large variance of the task performance. Previous literature (Devlin et al., 2019; Lee et al., 2020; Dodge et al., 2020) identified two potential reasons for the observed instability: catastrophic forgetting and small size of the fine-tuning datasets. In this paper, we show that both hypotheses fail to explain the fine-tuning instability. We analyze BERT, RoBERTa, and ALBERT, fine-tuned on commonly used datasets from the GLUE benchmark, and show that the observed instability is caused by optimization difficulties that lead to vanishing gradients. Additionally, we show that the remaining variance of the downstream task performance can be attributed to differences in generalization where fine-tuned models with the same training loss exhibit noticeably different test performance. Based on our analysis, we present a simple but strong baseline that makes fine-tuning BERT-based models significantly more stable than the previously proposed approaches. Code to reproduce our results is available online: https://github.com/uds-lsv/bert-stable-fine-tuning.
Multilingualism is a cultural cornerstone of Europe and firmly anchored in the European treaties including full language equality. However, language barriers impacting business, cross-lingual and cross-cultural communication are still omnipresent. Language Technologies (LTs) are a powerful means to break down these barriers. While the last decade has seen various initiatives that created a multitude of approaches and technologies tailored to Europe's specific needs, there is still an immense level of fragmentation. At the same time, AI has become an increasingly important concept in the European Information and Communication Technology area. For a few years now, AI, including many opportunities, synergies but also misconceptions, has been overshadowing every other topic. We present an overview of the European LT landscape, describing funding programmes, activities, actions and challenges in the different countries with regard to LT, including the current state of play in industry and the LT market. We present a brief overview of the main LT-related activities on the EU level in the last ten years and develop strategic guidance with regard to four key dimensions.
Teachable interfaces can empower end-users to attune machine learning systems to their idiosyncratic characteristics and environment by explicitly providing pertinent training examples. While facilitating control, their effectiveness can be hindered by the lack of expertise or misconceptions. We investigate how users may conceptualize, experience, and reflect on their engagement in machine teaching by deploying a mobile teachable testbed in Amazon Mechanical Turk. Using a performance-based payment scheme, Mechanical Turkers (N = 100) are called to train, test, and re-train a robust recognition model in real-time with a few snapshots taken in their environment. We find that participants incorporate diversity in their examples drawing from parallels to how humans recognize objects independent of size, viewpoint, location, and illumination. Many of their misconceptions relate to consistency and model capabilities for reasoning. With limited variation and edge cases in testing, the majority of them do not change strategies on a second training attempt.
Bayesian optimization (BO) is a popular approach to optimize expensive-to-evaluate black-box functions. A significant challenge in BO is to scale to high-dimensional parameter spaces while retaining sample efficiency. A solution considered in existing literature is to embed the high-dimensional space in a lower-dimensional manifold, often via a random linear embedding. In this paper, we identify several crucial issues and misconceptions about the use of linear embeddings for BO. We study the properties of linear embeddings from the literature and show that some of the design choices in current approaches adversely impact their performance. We show empirically that properly addressing these issues significantly improves the efficacy of linear embeddings for BO on a range of problems, including learning a gait policy for robot locomotion.