Course:CPSC522/Robot Scientist

From UBC Wiki
Jump to navigation Jump to search


Robot scientists are physically implemented laboratory automation systems that apply techniques from AI to automatically execute cycles of scientific experimentation.[1][2]

Paper 1 Automated Discovery in a Chemistry Laboratory

Paper 2 The Robot Scientist Adam

Principal Author: Ke Dai


Robot scientists are a natural development of the trend toward ever-greater computer involvment in science. They are physically implemented laboratory automation systems that apply techniques from AI to automatically execute cycles of scientific experimentation. They can automatically develop and test hypotheses to explain observations, run experiments using laboratory robotics, interpret the results to amend their hypotheses, and then repeat the cycle, automating high-throughput hypothesis-led research. This page presents two robot scientists which are applied in the domain of chemistry and molecular biology respecitively.


Evaluation and Comparison



Fahrenheit's performance can be evaluated according to accuracy, range of application, and usefulness. The accuracy of exprimental results by Fahrenheit is compatible with the accuracy achieved by human researchers. The range of applications of Fahrenheit goes far beyond electrochemistry, because Fahrenheit can deal with numerical data regardless of the application domain, provided that the laboratory hardware and the appropriate operational procedures are defined. In terms of usefulness, when comparing the theoretical work on the data made by Fahrenheit with an off-line analysis by the chemists, the results are very favorable. More than one day of work on data by a research assistant corresponds to 1 minute of data analysis by Fahrenheit. Altogether, several days work of research assistants corresponds to 50 minutes work of Fahrenheit.


Although Fahrenheit can automatically determine the repeatability conditions and it can find regularities for the peaks, the transition from the first to the second requires humans intervention. The corresponding change in operational procedure, however, has not been automated and must be done mannually.



Although Fahrenheit demonstrates the automation of experimantation, data collection, analysis and theory formation, it still requires human intervention at some steps of operational prodecure. However Adam fully close the loop in the sense of examining the experiments' results, deciding what to do next, and thus potentially cycling forever. Adam can run for a few days without human intervention in principle. Furthermore, Adam simulates the complete process of scientific research conducted by a real scientist by incorporating background knowledge, hypothesis generation, experiment selection and planning experiments module which do not exist in Fahrenheit. Typically, Adam can decide what experiments it executes while Fahrenheit can only repeat conducting the experiment prearranged by humans.

Annotated Bibliography

  1. R.D. King et al., “Functional Genomic Hypothesis Generation and Experimentation by a Robot Scientist,” Nature, vol. 427, 2004, pp. 247-252.
  2. R.D. King et al., “The Automation of Science,” Science, Apr. 2009, pp. 85-89.

To Add

Put links and content here to be added. This does not need to be organized, and will not be graded as part of the page. If you find something that might be useful for a page, feel free to put it here.

Some rights reserved
Permission is granted to copy, distribute and/or modify this document according to the terms in Creative Commons License, Attribution-NonCommercial-ShareAlike 3.0. The full text of this license may be found here: CC by-nc-sa 3.0