On Automating the Doctrine of Double Effect
Authors: Naveen Sundar Govindarajulu, Selmer Bringsjord
IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We then use our framework to successfully simulate scenarios that have been used to test for the presence of the principle in human subjects. Our framework can be used in two different modes: One can use it to build DDE-compliant autonomous systems from scratch; or one can use it to verify that a given AI system is DDE-compliant, by applying a DDE layer on an existing system or model. ... This is preliminary work to illustrate the feasibility of the second mode, and we hope that our initial sketches can be useful for other researchers in incorporating DDE in their own frameworks. ... We have achieved the first computational simulations of the doctrine. A byproduct of these simulations is an event-calculus formalization of a demanding class of trolley problems (widely used in empirical and philosophical studies of ethics). |
| Researcher Affiliation | Academia | Naveen Sundar Govindarajulu and Selmer Bringsjord Rensselaer Polytechnic Institute, Troy, NY {naveensundarg,selmer.bringsjord}@gmail.com |
| Pseudocode | No | The paper presents formal definitions (F1-F4) and inference schemata for the deontic cognitive event calculus (DCEC) but does not include any blocks explicitly labeled as 'Pseudocode' or 'Algorithm'. |
| Open Source Code | Yes | The prover is available in both Java and Common Lisp and can be obtained at: https://github.com/naveensundarg/prover. The underlying first-order prover is SNARK available at: http://www.ai.sri.com/ stickel/snark.html. |
| Open Datasets | No | The paper describes 'trolley problems' as scenarios used for simulation, but it does not refer to them as a publicly available dataset with specific access information (link, DOI, citation with author/year, or repository). |
| Dataset Splits | No | The paper does not provide specific details on dataset splits (e.g., percentages or sample counts) for training, validation, or testing. It describes scenarios rather than formal dataset partitions. |
| Hardware Specification | No | The paper describes the use of 'Shadow Prover' for simulations and provides run times. However, it does not specify any hardware details such as GPU models, CPU specifications, or memory used for the experiments. |
| Software Dependencies | No | The paper mentions using 'Shadow Prover' and that its underlying first-order prover is 'SNARK', and that the implementation uses 'Java and Common Lisp'. However, it does not provide specific version numbers for any of these software components, which is required for reproducibility. |
| Experiment Setup | Yes | We use a discrete version of the event calculus, in which time is discrete, but other quantities and measures, such as the utility function, can be continuous. We have the following additional sorts: Trolley and Track. We also declare that the Agent and the Trolley sorts are subsorts of a Moveable sort, the instances of which are objects that can be placed on tracks and moved. We use the following additional core symbols: ... The utility function µ is defined as follows: µ( f,t) = 1 if f dead(P) 0 otherwise We set the threshold γ at 0.5. The simulation starts at time t = 0... |