Related papers: Persistence Paradox in Dynamic Science

Over the last four decades, the way knowledge is created in academia has transformed dramatically: research teams have grown larger, scholars draw from ever-wider pools of prior work, and the most influential discoveries increasingly emerge…

In recent years, machine learning techniques have been increasingly applied in sensitive decision making processes, raising fairness concerns. Past research has shown that machine learning may reproduce and even exacerbate human bias due to…

One of the fundamental questions in science is how scientific disciplines evolve and sustain progress in society. No studies to date allows us to explain the endogenous processes that support the evolution of scientific disciplines and…

Robustness is often regarded as a critical future challenge for real-world applications, where stability is essential. However, as models often learn tasks in a similar order, we hypothesize that easier tasks will be easier regardless of…

Continuous adaptation allows survival in an ever-changing world. Adjustments in the synaptic coupling strength between neurons are essential for this capability, setting us apart from simpler, hard-wired organisms. How these changes can be…

The emergence of Deep Learning has marked a profound shift in machine learning, driven by numerous breakthroughs achieved in recent years. However, as Deep Learning becomes increasingly present in everyday tools and applications, there is a…

Theories of scientific and technological change view discovery and invention as endogenous processes, wherein prior accumulated knowledge enables future progress by allowing researchers to, in Newton's words, "stand on the shoulders of…

Technology enables a more sustainable and universally accessible educational model. However, technology has brought a paradox into students' lives: it helps them engage in learning activities, but it is also a source of distraction. During…

The selection of research topics by scientists can be viewed as an exploration process conducted by individuals with cognitive limitations traversing a complex cognitive landscape influenced by both individual and social factors. While…

In recent years deep reinforcement learning (RL) systems have attained superhuman performance in a number of challenging task domains. However, a major limitation of such applications is their demand for massive amounts of training data. A…

Machine learning and deep learning techniques are contributing much to the advancement of science. Their powerful predictive capabilities appear in numerous disciplines, including chaotic dynamics, but they miss understanding. The main…

Continual learning is the problem of learning and retaining knowledge through time over multiple tasks and environments. Research has primarily focused on the incremental classification setting, where new tasks/classes are added at discrete…

Disagreement is essential to scientific progress. However, the extent of disagreement in science, its evolution over time, and the fields in which it happens, remains poorly understood. Leveraging a massive collection of English-language…

Collectives form non-equilibrium social structures characterised by a volatile dynamics. Individuals join or leave. Social relations change quickly. Therefore, differently from engineered or ecological systems, a resilient reference state…

We examine the innovation of researchers with long-lived careers in Computer Science and Physics. Despite the epistemological differences between such disciplines, we consistently find that a researcher's most innovative publication occurs…

Time-dependent data-generating distributions have proven to be difficult for gradient-based training of neural networks, as the greedy updates result in catastrophic forgetting of previously learned knowledge. Despite the progress in the…

Student performance prediction - where a machine forecasts the future performance of students as they interact with online coursework - is a challenging problem. Reliable early-stage predictions of a student's future performance could be…

We approach productivity in science in a longitudinal fashion: We track careers over time, up to 40 years. We first allocate scientists to decile-based publishing productivity classes, from the bottom 10% to the top 10%. Then, we seek…

Due to the edge's position between the cloud and the users, and the recent surge of deep neural network (DNN) applications, edge computing brings about uncertainties that must be understood separately. Particularly, the edge users' locally…

The adaptive learning capabilities seen in biological neural networks are largely a product of the self-modifying behavior emerging from online plastic changes in synaptic connectivity. Current methods in Reinforcement Learning (RL) only…