LIVIVO - Das Suchportal für Lebenswissenschaften

switch to English language
Zurück zur letzten Suche Suchhistorie
Erweiterte Suche

Ihre letzten Suchen

  1. AU="Kirsch, L"
  2. AU="Yurong Qiao"
  3. AU="Shapera, Shane"
  4. AU="O'Connor, Richard J"
  5. AU="Li, Zhixing"
  6. AU="Fender, Christian"
  7. AU="Frangou, Nikoletta"
  8. AU="Chan, Curtis"
  9. AU="Yang, Shilun"
  10. AU="Viswanathan, Thiruselvam"
  11. AU="Rexach, Irene"
  12. AU="CUI Yongchun"

Suchergebnis

Treffer 1 - 10 von insgesamt 375

Suchoptionen

  1. Buch ; Online: Eliminating Meta Optimization Through Self-Referential Meta Learning

    Kirsch, Louis / Schmidhuber, Jürgen

    2022  

    Abstract: Meta Learning automates the search for learning algorithms. At the same time, it creates a dependency on human engineering on the meta-level, where meta learning algorithms need to be designed. In this paper, we investigate self-referential meta learning ...

    Abstract Meta Learning automates the search for learning algorithms. At the same time, it creates a dependency on human engineering on the meta-level, where meta learning algorithms need to be designed. In this paper, we investigate self-referential meta learning systems that modify themselves without the need for explicit meta optimization. We discuss the relationship of such systems to in-context and memory-based meta learning and show that self-referential neural networks require functionality to be reused in the form of parameter sharing. Finally, we propose fitness monotonic execution (FME), a simple approach to avoid explicit meta optimization. A neural network self-modifies to solve bandit and classic control tasks, improves its self-modifications, and learns how to learn, purely by assigning more computational resources to better performing solutions.

    Comment: The first version appeared at ICML 2022, DARL Workshop
    Schlagwörter Computer Science - Machine Learning ; Computer Science - Artificial Intelligence ; Computer Science - Neural and Evolutionary Computing ; Statistics - Machine Learning
    Thema/Rubrik (Code) 006
    Erscheinungsdatum 2022-12-29
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  2. Artikel ; Online: Perceptual heterogeneity in developmental prosopagnosia is continuous, not categorical.

    DeGutis, Joseph / Kirsch, Leah / Evans, Travis C / Fry, Regan / Lee, Daniel J / Mishra, Maruti / Campbell, Alison

    Cortex; a journal devoted to the study of the nervous system and behavior

    2024  Band 176, Seite(n) 37–52

    Abstract: Developmental prosopagnosia (DP) is associated with considerable perceptual heterogeneity, though the nature of this heterogeneity and whether there are discrete subgroups versus continuous deficits remains unclear. Bennetts et al. (2022) recently found ... ...

    Abstract Developmental prosopagnosia (DP) is associated with considerable perceptual heterogeneity, though the nature of this heterogeneity and whether there are discrete subgroups versus continuous deficits remains unclear. Bennetts et al. (2022) recently found that holistic versus featural processing deficits distinguished discrete DP subgroups, but their sample was relatively small (N = 37), and subgroups were defined using a single task. To characterize perceptual heterogeneity in DPs more comprehensively, we administered a broad face perception battery to a large sample of 109 DPs and 134 controls, including validated measures of face matching (Cambridge Face Perception Test - CFPT, Computerized Benton Facial Recognition Test, Same/Different Face Matching Task), holistic processing (Part-Whole Task), and feature processing (Georges Task and Part-Whole part trials). When examining face matching measures, DPs exhibited a similar distribution of performance as controls, though shifted towards impairment by an average of 1.4 SD. We next applied Bennetts (2022) hierarchical clustering approach and k-means clustering to the CFPT upright, inverted, and inversion index measures, similarly finding one group of DPs with poorer inverted face performance and another with a decreased face inversion effect (holistic processing). However, these subgroup differences failed to generalize to other measures of feature and holistic processing beyond the CFPT. We finally ran hierarchical and k-means cluster analyses on our larger battery of face matching, feature, and holistic processing measures. Results clearly showed subgroups with generally better versus worse performance across all measures, with the distinction between groups being somewhat arbitrary. Together, these findings support a continuous account of DP perceptual heterogeneity, with performance differing primarily across all aspects of face perception.
    Sprache Englisch
    Erscheinungsdatum 2024-04-25
    Erscheinungsland Italy
    Dokumenttyp Journal Article
    ZDB-ID 280622-8
    ISSN 1973-8102 ; 0010-9452
    ISSN (online) 1973-8102
    ISSN 0010-9452
    DOI 10.1016/j.cortex.2024.03.011
    Datenquelle MEDical Literature Analysis and Retrieval System OnLINE

    Zusatzmaterialien

    Kategorien

    Verfügbar in ZB MED Köln/Königswinter

    Zs.A 687: Hefte anzeigen Standort:
    Je nach Verfügbarkeit (siehe Angabe bei Bestand)
    bis Jg. 1994: Bestellungen von Artikeln über das Online-Bestellformular
    Jg. 1995 - 2021: Lesesall (1.OG)
    ab Jg. 2022: Lesesaal (EG)

    Über subito bestellen

    Dieser Service ist kostenpflichtig (siehe Lieferbedingungen von subito). Bestellungen, die einen Artikel nebst Supplementary Material umfassen, werden grundsätzlich wie mehrfache Bestellungen bearbeitet. Gebühren fallen in diesen Fällen für jede einzelne Bestellung an.

  3. Artikel: Embolie pulmonaire chez des patients atteints de COVID-19 : à propos de 6 cas.

    Steeman, A / Mazairac, G / Kirsch, L / Frusch, N / Morandini, E / Benoit, A

    Revue medicale de Liege

    2020  Band 75, Heft S1, Seite(n) 94–100

    Abstract: Rising from the province of Wuhan in China, the new coronavirus SARS-CoV-2 broke out in winter 2019, causing a global pandemic. In most cases reported, COVID-19 symptoms include cough, dyspnea, myalgia and asthenia. In some cases, the disease can also ... ...

    Titelübersetzung Six cases of acute pulmonary embolism associated with COVID-19.
    Abstract Rising from the province of Wuhan in China, the new coronavirus SARS-CoV-2 broke out in winter 2019, causing a global pandemic. In most cases reported, COVID-19 symptoms include cough, dyspnea, myalgia and asthenia. In some cases, the disease can also cause severe respiratory distress syndrome, requiring intensive care. Recent studies suggest that SARS-CoV-2 infection predisposes to thromboembolic event such as pulmonary embolism. Moreover, there is an overlap between signs and symptoms of pulmonary embolism and COVID-19, which brings a challenge for the diagnosis and could potentially be fatal. Nevertheless, the incidence rate of pulmonary embolism in cases of COVID-19 is currently not known. In this paper we describe six cases of pulmonary embolism associated with COVID-19.
    Mesh-Begriff(e) Betacoronavirus ; COVID-19 ; China ; Coronavirus Infections ; Humans ; Pandemics ; Pneumonia, Viral ; Pulmonary Embolism/diagnosis ; Pulmonary Embolism/etiology ; SARS-CoV-2
    Schlagwörter covid19
    Sprache Französisch
    Erscheinungsdatum 2020-11-19
    Erscheinungsland Belgium
    Dokumenttyp Case Reports
    ZDB-ID 414001-1
    ISSN 0370-629X ; 0035-3663
    ISSN 0370-629X ; 0035-3663
    Datenquelle MEDical Literature Analysis and Retrieval System OnLINE

    Zusatzmaterialien

    Kategorien

    Verfügbar in ZB MED Köln/Königswinter

    Zs.A 597: Hefte anzeigen Standort:
    Je nach Verfügbarkeit (siehe Angabe bei Bestand)
    bis Jg. 1994: Bestellungen von Artikeln über das Online-Bestellformular
    Jg. 1995 - 2021: Lesesall (1.OG)
    ab Jg. 2022: Lesesaal (EG)

    Über subito bestellen

    Dieser Service ist kostenpflichtig (siehe Lieferbedingungen von subito). Bestellungen, die einen Artikel nebst Supplementary Material umfassen, werden grundsätzlich wie mehrfache Bestellungen bearbeitet. Gebühren fallen in diesen Fällen für jede einzelne Bestellung an.

  4. Buch ; Online: Learning One Abstract Bit at a Time Through Self-Invented Experiments Encoded as Neural Networks

    Herrmann, Vincent / Kirsch, Louis / Schmidhuber, Jürgen

    2022  

    Abstract: There are two important things in science: (A) Finding answers to given questions, and (B) Coming up with good questions. Our artificial scientists not only learn to answer given questions, but also continually invent new questions, by proposing ... ...

    Abstract There are two important things in science: (A) Finding answers to given questions, and (B) Coming up with good questions. Our artificial scientists not only learn to answer given questions, but also continually invent new questions, by proposing hypotheses to be verified or falsified through potentially complex and time-consuming experiments, including thought experiments akin to those of mathematicians. While an artificial scientist expands its knowledge, it remains biased towards the simplest, least costly experiments that still have surprising outcomes, until they become boring. We present an empirical analysis of the automatic generation of interesting experiments. In the first setting, we investigate self-invented experiments in a reinforcement-providing environment and show that they lead to effective exploration. In the second setting, pure thought experiments are implemented as the weights of recurrent neural networks generated by a neural experiment generator. Initially interesting thought experiments may become boring over time.

    Comment: 20 pages, 6 figures
    Schlagwörter Computer Science - Machine Learning ; Computer Science - Artificial Intelligence
    Thema/Rubrik (Code) 501
    Erscheinungsdatum 2022-12-29
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  5. Buch ; Online: The Benefits of Model-Based Generalization in Reinforcement Learning

    Young, Kenny / Ramesh, Aditya / Kirsch, Louis / Schmidhuber, Jürgen

    2022  

    Abstract: Model-Based Reinforcement Learning (RL) is widely believed to have the potential to improve sample efficiency by allowing an agent to synthesize large amounts of imagined experience. Experience Replay (ER) can be considered a simple kind of model, which ... ...

    Abstract Model-Based Reinforcement Learning (RL) is widely believed to have the potential to improve sample efficiency by allowing an agent to synthesize large amounts of imagined experience. Experience Replay (ER) can be considered a simple kind of model, which has proved effective at improving the stability and efficiency of deep RL. In principle, a learned parametric model could improve on ER by generalizing from real experience to augment the dataset with additional plausible experience. However, given that learned value functions can also generalize, it is not immediately obvious why model generalization should be better. Here, we provide theoretical and empirical insight into when, and how, we can expect data generated by a learned model to be useful. First, we provide a simple theorem motivating how learning a model as an intermediate step can narrow down the set of possible value functions more than learning a value function directly from data using the Bellman equation. Second, we provide an illustrative example showing empirically how a similar effect occurs in a more concrete setting with neural network function approximation. Finally, we provide extensive experiments showing the benefit of model-based learning for online RL in environments with combinatorial complexity, but factored structure that allows a learned model to generalize. In these experiments, we take care to control for other factors in order to isolate, insofar as possible, the benefit of using experience generated by a learned model relative to ER alone.

    Comment: Update to ICML version
    Schlagwörter Computer Science - Machine Learning
    Thema/Rubrik (Code) 006
    Erscheinungsdatum 2022-11-03
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  6. Artikel ; Online: Empirical characterisation of agents’ spatial behaviour in pedestrian movement simulation

    Filomena, Gabriele / Kirsch, Lia / Schwering, Angela / Verstegen, Judith

    Journal of Environmental Psychology

    2022  Band 82

    Abstract: Route choice behaviour is an important factor in determining pedestrian movement flows across the urban space. Agent-based modelling, a simulation paradigm that allows modelling individual behaviour mechanisms to observe the emergence of macro-level ... ...

    Abstract Route choice behaviour is an important factor in determining pedestrian movement flows across the urban space. Agent-based modelling, a simulation paradigm that allows modelling individual behaviour mechanisms to observe the emergence of macro-level patterns, has not employed empirical data regarding route choice behaviour in cities or accommodated heterogeneity. The aim of this paper is to present an empirically based Agent-Based Model (ABM) that accounts for behavioural heterogeneity in pedestrian route choice strategies, for simulating the movement of pedestrians in cities. We designed a questionnaire to observe to what degree people resort to salient urban elements (i.e. local and global landmarks, regions, and barriers) and road costs (i.e. road distance, cumulative angular change) and empirically characterise the agent behaviour in our ABM. We hypothesised that a heterogeneous ABM configuration based on the construction of agent typologies from empirical data would portray a more plausible picture of pedestrian movement flows than a homogeneous configuration based on the same data, or a random configuration. The city of Münster (DE) was used as a case study. From a sample of 301 subjects, we obtained six clusters that differed in the usage of global elements (distant landmarks, barriers, and regions) and meaningful local elements along the route, when they choose their route. The random configuration directed agents nearby natural elements and across the streets of the historical centre. The empirically based model configurations resulted in lower pedestrian volumes along roads designed for cars (25% decrease) but higher concentrations along the city's Promenade and the lake (40% increase); based on our knowledge, we deem those results more plausible. Little differences were identified between the heterogeneous and homogeneous configurations. While these findings indicate that the inclusion of heterogeneity does not make a difference in terms of global patterns, we demonstrated that simulation models of ...
    Schlagwörter Life Science
    Thema/Rubrik (Code) 380
    Sprache Englisch
    Erscheinungsland nl
    Dokumenttyp Artikel ; Online
    ISSN 0272-4944
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Über subito bestellen

    Dieser Service ist kostenpflichtig (siehe Lieferbedingungen von subito). Bestellungen, die einen Artikel nebst Supplementary Material umfassen, werden grundsätzlich wie mehrfache Bestellungen bearbeitet. Gebühren fallen in diesen Fällen für jede einzelne Bestellung an.

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  7. Buch ; Online: Meta Learning Backpropagation And Improving It

    Kirsch, Louis / Schmidhuber, Jürgen

    2020  

    Abstract: Many concepts have been proposed for meta learning with neural networks (NNs), e.g., NNs that learn to reprogram fast weights, Hebbian plasticity, learned learning rules, and meta recurrent NNs. Our Variable Shared Meta Learning (VSML) unifies the above ... ...

    Abstract Many concepts have been proposed for meta learning with neural networks (NNs), e.g., NNs that learn to reprogram fast weights, Hebbian plasticity, learned learning rules, and meta recurrent NNs. Our Variable Shared Meta Learning (VSML) unifies the above and demonstrates that simple weight-sharing and sparsity in an NN is sufficient to express powerful learning algorithms (LAs) in a reusable fashion. A simple implementation of VSML where the weights of a neural network are replaced by tiny LSTMs allows for implementing the backpropagation LA solely by running in forward-mode. It can even meta learn new LAs that differ from online backpropagation and generalize to datasets outside of the meta training distribution without explicit gradient calculation. Introspection reveals that our meta learned LAs learn through fast association in a way that is qualitatively different from gradient descent.

    Comment: Updated to the NeurIPS 2021 camera ready; fixed typo in eq 4
    Schlagwörter Computer Science - Machine Learning ; Computer Science - Artificial Intelligence ; Computer Science - Neural and Evolutionary Computing ; Statistics - Machine Learning
    Thema/Rubrik (Code) 006
    Erscheinungsdatum 2020-12-29
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  8. Artikel ; Online: Downregulation of testicular function in the goat by altrenogest.

    Mihsler-Kirsch, Lisa / Wagner, Henrik / Failing, Klaus / Wehrend, Axel

    BMC veterinary research

    2021  Band 17, Heft 1, Seite(n) 183

    Abstract: Background: The present study investigated whether the administration of the progestin altrenogest provides noninvasive, temporary, and reversible suppression of gonadal function in the goat as a potential alternative to chirurgical castration, which is ...

    Abstract Background: The present study investigated whether the administration of the progestin altrenogest provides noninvasive, temporary, and reversible suppression of gonadal function in the goat as a potential alternative to chirurgical castration, which is related with irreversibility, risks of complications till death of the animal and welfare issues. Eight sexually mature Peacock goats were randomly divided into two groups. The experimental group was administered altrenogest (0.088 mg/kg) orally once daily for 7 weeks. The remaining four goats received an oral glucose solution and served as the control group. After completing the administration period, the reversibility of the medication was evaluated for another 7 weeks (observation phase). The treatment effects were assessed by clinical examination; ultrasound examination of the testes, including one-dimensional grayscale analysis, blood testosterone levels, analysis of semen parameters and libido. At the end of the observation period, the animals were castrated and the testicles were examined histologically.
    Results: Altrenogest treatment had no significant effect on the physical development of the goats, the sonographic appearance of the testes, the gray values measured in the ultrasound images, or the blood testosterone levels. The effects of treatment on the testicular and semen parameters varied widely in the experimental animals; the testicle volume was significantly lower and the number of pathologically altered sperm in the ejaculate was significantly higher in treated animals.
    Conclusion: These findings indicate that daily altrenogest administration at a dose of 0.088 mg/kg does not reliably suppress gonadal function in the goat.
    Mesh-Begriff(e) Administration, Oral ; Animals ; Contraceptive Agents, Male/administration & dosage ; Contraceptive Agents, Male/pharmacology ; Goats ; Male ; Semen Analysis/veterinary ; Sexual Behavior, Animal/drug effects ; Testis/diagnostic imaging ; Testis/drug effects ; Testosterone/blood ; Trenbolone Acetate/administration & dosage ; Trenbolone Acetate/analogs & derivatives ; Trenbolone Acetate/pharmacology
    Chemische Substanzen Contraceptive Agents, Male ; altrenogest (2U0X0JA2NB) ; Testosterone (3XMK78S47O) ; Trenbolone Acetate (RUD5Y4SV0S)
    Sprache Englisch
    Erscheinungsdatum 2021-05-04
    Erscheinungsland England
    Dokumenttyp Journal Article
    ISSN 1746-6148
    ISSN (online) 1746-6148
    DOI 10.1186/s12917-021-02845-6
    Datenquelle MEDical Literature Analysis and Retrieval System OnLINE

    Zusatzmaterialien

    Kategorien

    Über subito bestellen

    Dieser Service ist kostenpflichtig (siehe Lieferbedingungen von subito). Bestellungen, die einen Artikel nebst Supplementary Material umfassen, werden grundsätzlich wie mehrfache Bestellungen bearbeitet. Gebühren fallen in diesen Fällen für jede einzelne Bestellung an.

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  9. Buch ; Online: Goal-Conditioned Generators of Deep Policies

    Faccio, Francesco / Herrmann, Vincent / Ramesh, Aditya / Kirsch, Louis / Schmidhuber, Jürgen

    2022  

    Abstract: Goal-conditioned Reinforcement Learning (RL) aims at learning optimal policies, given goals encoded in special command inputs. Here we study goal-conditioned neural nets (NNs) that learn to generate deep NN policies in form of context-specific weight ... ...

    Abstract Goal-conditioned Reinforcement Learning (RL) aims at learning optimal policies, given goals encoded in special command inputs. Here we study goal-conditioned neural nets (NNs) that learn to generate deep NN policies in form of context-specific weight matrices, similar to Fast Weight Programmers and other methods from the 1990s. Using context commands of the form "generate a policy that achieves a desired expected return," our NN generators combine powerful exploration of parameter space with generalization across commands to iteratively find better and better policies. A form of weight-sharing HyperNetworks and policy embeddings scales our method to generate deep NNs. Experiments show how a single learned policy generator can produce policies that achieve any return seen during training. Finally, we evaluate our algorithm on a set of continuous control tasks where it exhibits competitive performance. Our code is public.

    Comment: Preprint. Under Review
    Schlagwörter Computer Science - Machine Learning ; Statistics - Machine Learning
    Thema/Rubrik (Code) 006
    Erscheinungsdatum 2022-07-04
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  10. Buch ; Online: General-Purpose In-Context Learning by Meta-Learning Transformers

    Kirsch, Louis / Harrison, James / Sohl-Dickstein, Jascha / Metz, Luke

    2022  

    Abstract: Modern machine learning requires system designers to specify aspects of the learning pipeline, such as losses, architectures, and optimizers. Meta-learning, or learning-to-learn, instead aims to learn those aspects, and promises to unlock greater ... ...

    Abstract Modern machine learning requires system designers to specify aspects of the learning pipeline, such as losses, architectures, and optimizers. Meta-learning, or learning-to-learn, instead aims to learn those aspects, and promises to unlock greater capabilities with less manual effort. One particularly ambitious goal of meta-learning is to train general-purpose in-context learning algorithms from scratch, using only black-box models with minimal inductive bias. Such a model takes in training data, and produces test-set predictions across a wide range of problems, without any explicit definition of an inference model, training loss, or optimization algorithm. In this paper we show that Transformers and other black-box models can be meta-trained to act as general-purpose in-context learners. We characterize transitions between algorithms that generalize, algorithms that memorize, and algorithms that fail to meta-train at all, induced by changes in model size, number of tasks, and meta-optimization. We further show that the capabilities of meta-trained algorithms are bottlenecked by the accessible state size (memory) determining the next prediction, unlike standard models which are thought to be bottlenecked by parameter count. Finally, we propose practical interventions such as biasing the training distribution that improve the meta-training and meta-generalization of general-purpose in-context learning algorithms.

    Comment: Published at the NeurIPS 2022 Workshop on Meta-Learning. Full version currently under review
    Schlagwörter Computer Science - Machine Learning ; Computer Science - Artificial Intelligence ; Computer Science - Neural and Evolutionary Computing ; Statistics - Machine Learning
    Thema/Rubrik (Code) 006
    Erscheinungsdatum 2022-12-08
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

Zum Seitenanfang