The Experts below are selected from a list of 87 Experts worldwide ranked by ideXlab platform
Michèle Sebag - One of the best experts on this subject based on the ideXlab platform.
-
Relational learning as search in a critical region
Journal of Machine Learning Research, 2003Co-Authors: Marco Botta, Attilio Giordana, Lorenza Saitta, Michèle SebagAbstract:Machine learning strongly relies on the covering test to assess whether a Candidate Hypothesis covers training examples. The present paper investigates learning relational concepts from examples, termed relational learning or inductive logic programming. In particular, it investigates the chances of success and the computational cost of relational learning, which appears to be severely affected by the presence of a phase transition in the covering test. To this aim, three up-to-date relational learners have been applied to a wide range of artificial, fully relational learning problems. A first experimental observation is that the phase transition behaves as an attractor for relational learning; no matter which region the learning problem belongs to, all three learners produce hypotheses lying within or close to the phase transition region. Second, a failure region appears. All three learners fail to learn any accurate Hypothesis in this region. Quite surprisingly, the probability of failure does not systematically increase with the size of the underlying target concept: under some circumstances, longer concepts may be easier to accurately approximate than shorter ones. Some interpretations for these findings are proposed and discussed.
-
IJCAI (2) - Tractable induction and classification in first order logic via stochastic matching
1997Co-Authors: Michèle Sebag, Céline RouveirolAbstract:Learning in first-order logic (FOL) languages suffers from a specific difficulty: both induction and classification are potentially exponential in the size of hypotheses. This difficulty is usually dealt with by limiting the size of hypotheses, via either syntactic restrictions or search strategies. This paper is concerned with polynomial induction and use of FOL hypotheses with no size restrictions. This is done via stochastic matching: instead of exhaustively exploring the set of matchings between any example and any short Candidate Hypothesis, one stochastically explores the set of matchings between any example and any Candidate Hypothesis. The user sets the number of matching samples to consider and thereby controls the cost of induction and classification. One advantage of this heuristic is to allow for resource-bounded learning, without any a priori knowledge about the problem domain. Experiments on a real-world problem pertaining to organic chemistry fully demonstrate the potentialities of the approach regarding both predictive accuracy and computational cost.
-
Inductive Logic Programming Workshop - Polynomial-Time Learning in Logic Programming and Constraint Logic Programming
Inductive Logic Programming, 1997Co-Authors: Michèle Sebag, Céline RouveirolAbstract:Induction in first-order logic languages suffers from an additional factor of complexity compared to induction in attribute-value languages: the number of possible matchings between a Candidate Hypothesis and a training example.
Michael Williams - One of the best experts on this subject based on the ideXlab platform.
-
replicated replicable and relevant target engagement and pharmacological experimentation in the 21st century
Biochemical Pharmacology, 2014Co-Authors: Terry P Kenakin, David B Bylund, Myron L Toews, Kevin Mullane, Raymond J Winquist, Michael WilliamsAbstract:Abstract A pharmacological experiment is typically conducted to: i) test or expand a Hypothesis regarding the potential role of a target in the mechanism(s) underlying a disease state using an existing drug or tool compound in normal and/or diseased tissue or animals; or ii) characterize and optimize a new chemical entity (NCE) targeted to modulate a specific disease-associated target to restore homeostasis as a potential drug Candidate. Hypothesis testing necessitates an intellectually rigorous, null Hypothesis approach that is distinct from a high throughput fishing expedition in search of a Hypothesis. In conducting an experiment, the protocol should be transparently defined along with its powering, design, appropriate statistical analysis and consideration of the anticipated outcome (s) before it is initiated. Compound-target interactions often involve the direct study of phenotype(s) unique to the target at the cell, tissue or animal/human level. However, in vivo studies are often compromised by a lack of sufficient information on the compound pharmacokinetics necessary to ensure target engagement and also by the context-free analysis of ubiquitous cellular signaling pathways downstream from the target. The use of single tool compounds/drugs at one concentration in engineered cell lines frequently results in reductionistic data that have no physiologically relevance. This overview, focused on trends in the peer-reviewed literature, discusses the execution and reporting of experiments and the criteria recommended for the physiologically-relevant assessment of target engagement to identify viable new drug targets and facilitate the advancement of translational studies.
Céline Rouveirol - One of the best experts on this subject based on the ideXlab platform.
-
IJCAI (2) - Tractable induction and classification in first order logic via stochastic matching
1997Co-Authors: Michèle Sebag, Céline RouveirolAbstract:Learning in first-order logic (FOL) languages suffers from a specific difficulty: both induction and classification are potentially exponential in the size of hypotheses. This difficulty is usually dealt with by limiting the size of hypotheses, via either syntactic restrictions or search strategies. This paper is concerned with polynomial induction and use of FOL hypotheses with no size restrictions. This is done via stochastic matching: instead of exhaustively exploring the set of matchings between any example and any short Candidate Hypothesis, one stochastically explores the set of matchings between any example and any Candidate Hypothesis. The user sets the number of matching samples to consider and thereby controls the cost of induction and classification. One advantage of this heuristic is to allow for resource-bounded learning, without any a priori knowledge about the problem domain. Experiments on a real-world problem pertaining to organic chemistry fully demonstrate the potentialities of the approach regarding both predictive accuracy and computational cost.
-
Inductive Logic Programming Workshop - Polynomial-Time Learning in Logic Programming and Constraint Logic Programming
Inductive Logic Programming, 1997Co-Authors: Michèle Sebag, Céline RouveirolAbstract:Induction in first-order logic languages suffers from an additional factor of complexity compared to induction in attribute-value languages: the number of possible matchings between a Candidate Hypothesis and a training example.
Terry P Kenakin - One of the best experts on this subject based on the ideXlab platform.
-
replicated replicable and relevant target engagement and pharmacological experimentation in the 21st century
Biochemical Pharmacology, 2014Co-Authors: Terry P Kenakin, David B Bylund, Myron L Toews, Kevin Mullane, Raymond J Winquist, Michael WilliamsAbstract:Abstract A pharmacological experiment is typically conducted to: i) test or expand a Hypothesis regarding the potential role of a target in the mechanism(s) underlying a disease state using an existing drug or tool compound in normal and/or diseased tissue or animals; or ii) characterize and optimize a new chemical entity (NCE) targeted to modulate a specific disease-associated target to restore homeostasis as a potential drug Candidate. Hypothesis testing necessitates an intellectually rigorous, null Hypothesis approach that is distinct from a high throughput fishing expedition in search of a Hypothesis. In conducting an experiment, the protocol should be transparently defined along with its powering, design, appropriate statistical analysis and consideration of the anticipated outcome (s) before it is initiated. Compound-target interactions often involve the direct study of phenotype(s) unique to the target at the cell, tissue or animal/human level. However, in vivo studies are often compromised by a lack of sufficient information on the compound pharmacokinetics necessary to ensure target engagement and also by the context-free analysis of ubiquitous cellular signaling pathways downstream from the target. The use of single tool compounds/drugs at one concentration in engineered cell lines frequently results in reductionistic data that have no physiologically relevance. This overview, focused on trends in the peer-reviewed literature, discusses the execution and reporting of experiments and the criteria recommended for the physiologically-relevant assessment of target engagement to identify viable new drug targets and facilitate the advancement of translational studies.
David B Bylund - One of the best experts on this subject based on the ideXlab platform.
-
replicated replicable and relevant target engagement and pharmacological experimentation in the 21st century
Biochemical Pharmacology, 2014Co-Authors: Terry P Kenakin, David B Bylund, Myron L Toews, Kevin Mullane, Raymond J Winquist, Michael WilliamsAbstract:Abstract A pharmacological experiment is typically conducted to: i) test or expand a Hypothesis regarding the potential role of a target in the mechanism(s) underlying a disease state using an existing drug or tool compound in normal and/or diseased tissue or animals; or ii) characterize and optimize a new chemical entity (NCE) targeted to modulate a specific disease-associated target to restore homeostasis as a potential drug Candidate. Hypothesis testing necessitates an intellectually rigorous, null Hypothesis approach that is distinct from a high throughput fishing expedition in search of a Hypothesis. In conducting an experiment, the protocol should be transparently defined along with its powering, design, appropriate statistical analysis and consideration of the anticipated outcome (s) before it is initiated. Compound-target interactions often involve the direct study of phenotype(s) unique to the target at the cell, tissue or animal/human level. However, in vivo studies are often compromised by a lack of sufficient information on the compound pharmacokinetics necessary to ensure target engagement and also by the context-free analysis of ubiquitous cellular signaling pathways downstream from the target. The use of single tool compounds/drugs at one concentration in engineered cell lines frequently results in reductionistic data that have no physiologically relevance. This overview, focused on trends in the peer-reviewed literature, discusses the execution and reporting of experiments and the criteria recommended for the physiologically-relevant assessment of target engagement to identify viable new drug targets and facilitate the advancement of translational studies.
