A Language for Counterfactual Generative Models
Authors: Zenna Tavares, James Koppel, Xin Zhang, Ria Das, Armando Solar-Lezama
ICML 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We accomplish this by introducing a new operator to probabilistic programming akin to Pearl s do, define its formal semantics, provide an implementation, and demonstrate its utility through examples in a variety of simulation models.Here we demonstrate counterfactual reasoning in OMEGAC through three case studies. All experiments were performed using predicate exchange (31). |
| Researcher Affiliation | Academia | Zenna Tavares 1 James Koppel 1 Xin Zhang 2 Ria Das 1 Armando Solar-Lezama 1 ... 1CSAIL, MIT, USA 2Key Lab of High Confidence Software Technologies, Ministry of Education, Department of Computer Science and Technology, Peking University, China. |
| Pseudocode | No | The paper provides abstract syntax for λC (Figures 2, 3, 4) and inline code examples to illustrate language features and concepts, but it does not include a distinct block explicitly labeled 'Pseudocode' or 'Algorithm'. |
| Open Source Code | Yes | A Julia implementation of OMEGAC can be found at https://github.com/zenna/Omega.jl, and a Haskell implementation of λC can be found at https://github.com/jkoppel/omega-calculus. |
| Open Datasets | Yes | Glucose Modelling This example queries whether a hypoglycemic episode could have been avoided in a diabetic patient. We first construct an ODE over variables captured in the Ohio Glucose dataset (17): (1) CGM: continuously monitored glucose measurements, (2) Steps: steps walked by patient, (3) Bolus: insulin injection events, and (4) Meals: calorie intake. |
| Dataset Splits | No | The paper focuses on demonstrating the capabilities of the OMEGAC language through case studies and examples, but it does not specify details regarding training, validation, or test dataset splits for reproducibility. |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., CPU, GPU models, memory) used to run the experiments described in the case studies. |
| Software Dependencies | No | The paper mentions that OMEGAC is implemented in Julia and λC in Haskell, and that 'All experiments were performed using predicate exchange (31)', but it does not specify exact version numbers for Julia, Haskell, or any other software dependencies. |
| Experiment Setup | No | The paper describes the setup of its simulation models for the case studies (e.g., initial scenes for car crash, time parameters for glucose model), but it does not provide specific hyperparameter values, optimizer settings, or other detailed system-level training configurations for model training. |