ILLC-LIRa March 2022

illc-lira@list.uva.nl

1 participants
7 discussions

LIRa Session: Bonan Zhao (tomorrow, Thursday, 24 March 16:30)
by LIRa Seminar 23 Mar '22

23 Mar '22

Dear all, We will have our next LIRa session tomorrow, on Thursday, 24 March 16:30. Please use our recurring zoom link: https://uva-live.zoom.us/j/88142993494?pwd=d1BsQWR4T2UyK0Job29YNThjaGRkUT09 (Meeting ID: 881 4299 3494, Passcode: 352984) You can find the details of the talk below. Speaker: Bonan Zhao (University of Edinburgh) Date and Time: Thursday, March 24th 2022, 16:30-18:00, Amsterdam time. Venue: online. Title: How do people generalize causal relations over objects? Abstract. How do people decide how general a causal relationship is, in terms of the entities or situations it applies to? What features do people use to decide whether a new situation is governed by a new causal law or an old one? How can people make these difficult judgments in a fast, efficient way? We address these questions in two experiments that ask participants to generalize from one (Experiment 1) or several (Experiment 2) causal interactions between pairs of objects. In each case, participants see an agent object act on a recipient object, causing some changes to the recipient. In line with the human capacity for few-shot concept learning, we find systematic patterns of causal generalizations favoring simpler causal laws that extend over categories of similar objects. In Experiment 1, we find that participants’ inferences are shaped by the order of the generalization questions they are asked. In both experiments, we find an asymmetry in the formation of causal categories: participants preferentially identify causal laws with features of the agent objects rather than recipients. To explain this, we develop a computational model that combines program induction (about the hidden causal laws) with non-parametric category inference (about their domains of influence). We demonstrate that our modeling approach can both explain the order effect in Experiment 1 and the causal asymmetry, and outperforms a naïve Bayesian account while providing a computationally plausible mechanism for real-world causal generalization. Hope to see you there! The LIRa team

1 0

LIRa Session: Bonan Zhao (Thursday, 24 March 16:30)
by LIRa Seminar 18 Mar '22

18 Mar '22

Dear all, We will have our next LIRa session on Thursday, 24 March 16:30. Please use our recurring zoom link: https://uva-live.zoom.us/j/88142993494?pwd=d1BsQWR4T2UyK0Job29YNThjaGRkUT09 (Meeting ID: 881 4299 3494, Passcode: 352984) You can find the details of the talk below. Speaker: Bonan Zhao (University of Edinburgh) Date and Time: Thursday, March 24th 2022, 16:30-18:00, Amsterdam time. Venue: online. Title: How do people generalize causal relations over objects? Abstract. How do people decide how general a causal relationship is, in terms of the entities or situations it applies to? What features do people use to decide whether a new situation is governed by a new causal law or an old one? How can people make these difficult judgments in a fast, efficient way? We address these questions in two experiments that ask participants to generalize from one (Experiment 1) or several (Experiment 2) causal interactions between pairs of objects. In each case, participants see an agent object act on a recipient object, causing some changes to the recipient. In line with the human capacity for few-shot concept learning, we find systematic patterns of causal generalizations favoring simpler causal laws that extend over categories of similar objects. In Experiment 1, we find that participants’ inferences are shaped by the order of the generalization questions they are asked. In both experiments, we find an asymmetry in the formation of causal categories: participants preferentially identify causal laws with features of the agent objects rather than recipients. To explain this, we develop a computational model that combines program induction (about the hidden causal laws) with non-parametric category inference (about their domains of influence). We demonstrate that our modeling approach can both explain the order effect in Experiment 1 and the causal asymmetry, and outperforms a naïve Bayesian account while providing a computationally plausible mechanism for real-world causal generalization. Hope to see you there! The LIRa team

1 0

LIRa session: Hans van Ditmarsch (tomorrow, Thursday, 17 March 16:30)
by LIRa Seminar 16 Mar '22

16 Mar '22

Dear all, We will have our next LIRa session tomorrow, on Thursday, 17 March 16:30. Please use our recurring zoom link: https://uva-live.zoom.us/j/88142993494?pwd=d1BsQWR4T2UyK0Job29YNThjaGRkUT09 (Meeting ID: 881 4299 3494, Passcode: 352984) You can find the details of the talk below. Speaker: Hans van Ditmarsch (Open University, Heerlen) Date and Time: Thursday, March 17th 2022, 16:30-18:00, Amsterdam time. Venue: online. Title: Wanted Dead or Alive: Epistemic logic for impure simplicial complexes Abstract. We propose a logic of knowledge for impure simplicial complexes. Impure simplicial complexes represent distributed systems under uncertainty over which processes are still active (are alive) and which processes have failed or crashed (are dead). Our work generalizes the logic of knowledge for pure simplicial complexes, where all processes are alive, by Goubault et al. Our logical semantics has a satisfaction relation defined simultaneously with a definability relation. The latter restricts which formulas are allowed to have a truth value: dead processes cannot know or be ignorant of any proposition, and live processes cannot know or be ignorant of propositions involving processes they know to be dead. The logic satisfies some but not all axioms and rules of the modal logic S5. Impure simplicial complexes correspond to Kripke models where each agent's accessibility relation is an equivalence relation on a subset of the domain only, and otherwise empty, and where each propositional variable is known by an agent. See also https://arxiv.org/abs/2103.03032 Hope to see you there! The LIRa team

1 0

LIRa session: Hans van Ditmarsch (Thursday, 17 March 16:30)
by LIRa Seminar 11 Mar '22

11 Mar '22

Dear all, We will have our next LIRa session on Thursday, 17 March 16:30. Please use our recurring zoom link: https://uva-live.zoom.us/j/88142993494?pwd=d1BsQWR4T2UyK0Job29YNThjaGRkUT09 (Meeting ID: 881 4299 3494, Passcode: 352984) You can find the details of the talk below. Speaker: Hans van Ditmarsch (Open University, Heerlen) Date and Time: Thursday, March 17th 2022, 16:30-18:00, Amsterdam time. Venue: online. Title: Wanted Dead or Alive: Epistemic logic for impure simplicial complexes Abstract. We propose a logic of knowledge for impure simplicial complexes. Impure simplicial complexes represent distributed systems under uncertainty over which processes are still active (are alive) and which processes have failed or crashed (are dead). Our work generalizes the logic of knowledge for pure simplicial complexes, where all processes are alive, by Goubault et al. Our logical semantics has a satisfaction relation defined simultaneously with a definability relation. The latter restricts which formulas are allowed to have a truth value: dead processes cannot know or be ignorant of any proposition, and live processes cannot know or be ignorant of propositions involving processes they know to be dead. The logic satisfies some but not all axioms and rules of the modal logic S5. Impure simplicial complexes correspond to Kripke models where each agent's accessibility relation is an equivalence relation on a subset of the domain only, and otherwise empty, and where each propositional variable is known by an agent. We also propose a notion of bisimulation for impure simplexes and show bisimulation correspondence on certain finitary simplexes. See also https://arxiv.org/abs/2103.03032 Hope to see you there! The LIRa team

1 0

LIRa session: Stipe Pandzic (tomorrow, Thursday, 10 March 16:30)
by LIRa Seminar 09 Mar '22

09 Mar '22

Dear all, We will have our next LIRa session tomorrow, on Thursday, 10 March 16:30. Please use our recurring zoom link: https://uva-live.zoom.us/j/88142993494?pwd=d1BsQWR4T2UyK0Job29YNThjaGRkUT09 (Meeting ID: 881 4299 3494, Passcode: 352984) You can find the details of the talk below. Speaker: Stipe Pandzic (Utrecht University) Date and Time: Thursday, March 10th 2022, 16:30-18:00, Amsterdam time. Venue: online. Title: Default justification logic as a theory of arguments and epistemic justification. Abstract. My goal in this talk is to present a logic of structured defeasible argumentation using the language of justification logic. One of the key features that is absent in standard justification logics is the possibility to weigh different epistemic reasons that might conflict with one another. To amend this, we develop a semantics for “defeaters”: conflicting reasons forming a basis to doubt the original conclusion or to believe an opposite statement. Formally, non-monotonicity of defeasible reasons is introduced through default rules with justification logic formulas. Default justification logic is unique for its ability to represent defeasible arguments as object-level formulas. Sets of such formulas are then interpreted with an acceptance semantics, in analogy to Dung’s abstract argumentation framework semantics. In contrast to argumentation frameworks, determining arguments’ acceptance in default justification logic simply turns into finding (non-monotonic) logical consequences from a starting theory with justification assertions. We can show that a large subclass of Dung’s frameworks is a special case of default justification logic. By the end of this talk, I will connect this logic of arguments with the epistemological debate on justified-true-belief definition of knowledge. Hope to see you there! The LIRa team

1 0

LIRa session: Stipe Pandzic (Thursday, 10 March 16:30)
by LIRa Seminar 04 Mar '22

04 Mar '22

Dear all, We will have our next LIRa session on Thursday, 10 March 16:30. Please use our recurring zoom link: https://uva-live.zoom.us/j/88142993494?pwd=d1BsQWR4T2UyK0Job29YNThjaGRkUT09 (Meeting ID: 881 4299 3494, Passcode: 352984) You can find the details of the talk below. Speaker: Stipe Pandzic (Utrecht University) Date and Time: Thursday, March 10th 2022, 16:30-18:00, Amsterdam time. Venue: online. Title: Default justification logic as a theory of arguments and epistemic justification. Abstract. My goal in this talk is to present a logic of structured defeasible argumentation using the language of justification logic. One of the key features that is absent in standard justification logics is the possibility to weigh different epistemic reasons that might conflict with one another. To amend this, we develop a semantics for “defeaters”: conflicting reasons forming a basis to doubt the original conclusion or to believe an opposite statement. Formally, non-monotonicity of defeasible reasons is introduced through default rules with justification logic formulas. Default justification logic is unique for its ability to represent defeasible arguments as object-level formulas. Sets of such formulas are then interpreted with an acceptance semantics, in analogy to Dung’s abstract argumentation framework semantics. In contrast to argumentation frameworks, determining arguments’ acceptance in default justification logic simply turns into finding (non-monotonic) logical consequences from a starting theory with justification assertions. We can show that a large subclass of Dung’s frameworks is a special case of default justification logic. By the end of this talk, I will connect this logic of arguments with the epistemological debate on justified-true-belief definition of knowledge. Hope to see you there! The LIRa team

1 0

LIRa Session: Louwe Kuijer (tomorrow, Thursday, 3 March 16:30)
by LIRa Seminar 02 Mar '22

02 Mar '22

Dear all, We will have our next LIRa session tomorrow, on Thursday, 3 March 16:30. Please use our recurring zoom link: https://uva-live.zoom.us/j/88142993494?pwd=d1BsQWR4T2UyK0Job29YNThjaGRkUT09 (Meeting ID: 881 4299 3494, Passcode: 352984) You can find the details of the talk below. Speaker: Louwe Kuijer (University of Liverpool) Date and Time: Thursday, March 3rd 2022, 16:30-18:00, Amsterdam time. Venue: online. Title: Doing the best I can. Abstract. A preference structure is a graph that indicates preferences between alternatives. These alternatives could be possible worlds, outcomes of an action, strategies in a game or something else entirely. The preferences, similarly, can mean many different things: a world might be preferred over another if it is more plausible, an outcome if it improves social welfare, and a strategy might be preferred over another strategy that it defeats. >From these preferences we can determine the set of "best" worlds, outcomes and strategies. Generally, one should believe the best worlds, strive to obtain the best outcomes and play the best strategies. As usual in deontic logic, we can then define a notion of obligation (to believe, bring about or play), where ϕ is obligatory, denoted B(ϕ) if all of the best alternatives satisfy ϕ (and thus it is impossible to satisfy the obligation to be "best'' without making ϕ true). This notion of obligation can be extended to a conditional one, B(ϕ|ψ), where ϕ is obligatory given ψ, by relativization. In other words, we restrict the preference structure to those alternatives that satisfy ψ, and check whether ϕ is true in all of the best alternatives in that structure. The properties of this conditional obligation operator depend on (1) exactly how we define what the "best" states in a preference structure are and (2) what properties the preference relation satisfies (e.g., transitive, total, anti-symmetric). In this talk, I will discuss characterizations of the conditional belief operator under various such assumptions. We will also consider one case where, despite searching for a long time, I did not manage to find a characterization: this turned out to be because such a characterization does not exist. Hope to see you there! The LIRa team

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

ILLC-LIRa March 2022