Results 41 to 50 of about 259,517 (255)
Personalizing Dialogue Agents via Meta-Learning
, 2019 Existing personalized dialogue models use human designed persona descriptions to improve dialogue consistency. Collecting such descriptions from existing dialogues is expensive and requires hand-crafted feature designs.
Fung, Pascale +3 more
core +1 more source
Meta Learning for Causal Direction
Proceedings of the AAAI Conference on Artificial Intelligence, 2021 The inaccessibility of controlled randomized trials due to inherent constraints in many fields of science has been a fundamental issue in causal inference. In this paper, we focus on distinguishing the cause from effect in the bivariate setting under limited observational data.
Ton, J-F, Sejdinovic, D, Fukumizu, K
openaire +3 more sources
Overview of molecular signatures of senescence and associated resources: pros and cons
FEBS Open Bio, EarlyView.Cells can enter a stress response state termed cellular senescence that is involved in various diseases and aging. Detecting these cells is challenging due to the lack of universal biomarkers. This review presents the current state of senescence identification, from biomarkers to molecular signatures, compares tools and approaches, and highlights ...
Orestis A. Ntintas +6 more
wiley +1 more source
BMC Health Services ResearchBackground Since the inception of the ASHAs in the year 2005, their work horizons have increased from Reproductive, Maternal, Newborn, Child, and Adolescent health (RMNCH + A), Communicable and Non-Communicable Diseases (CD & NCD) to oral health ...
Shambhavi Singh +7 more
doaj +1 more source
BMJ OpenObjectives This study aims to examine the prevalence of comparisons of surgery to drug regimens, the strength of evidence of such comparisons and whether surgery or the drug intervention was favoured.Design Systematic review of systematic reviews ...
Ewoud Schuit +5 more
doaj +1 more source
Few Shot Class Incremental Learning via Efficient Prototype Replay and Calibration
Entropy, 2023 Few shot class incremental learning (FSCIL) is an extremely challenging but valuable problem in real-world applications. When faced with novel few shot tasks in each incremental stage, it should take into account both catastrophic forgetting of old ...
Wei Zhang, Xiaodong Gu
doaj +1 more source
Towards Meta-learning over Data Streams [PDF]
, 2014 Modern society produces vast streams of data. Many stream mining algorithms have been developed to capture general trends in these streams, and make predictions for future observations, but relatively little is known about which algorithms perform ...
Holmes, Geoffrey +3 more
core +1 more source
FEBS Open Bio, EarlyView.What factors make for an effective digital learning tool in Higher Education? This systematic review identifies elements of a digital tool that published examples reveal to be features of an engaging and impactful digital tool. A systematic literature search yielded 25 research papers for analysis.
Akmal Arzeman +4 more
wiley +1 more source
IEEE AccessIntegrating knowledge across different domains is an essential feature of human learning. Learning paradigms such as transfer learning, meta-learning, and multi-task learning reflect the human learning process by exploiting the prior knowledge for new ...
Richa Upadhyay +3 more
doaj +1 more source
Remote Sensing, 2021 The central goal of few-shot scene classification is to learn a model that can generalize well to a novel scene category (UNSEEN) from only one or a few labeled examples.
Pei Zhang +4 more
doaj +1 more source