Israel
November 1, 2022
Developing Trust in Healthcare AI
November 1, 20221971-1973
1970-2000: The period between 1970 and 2000 is referred to as the “AI winter,” when artificial intelligence funding and interests were low. Many recognize two significant winters: the first in the late 1970s, fueled by perceptions of AI’s limits, and the second winter in the late 1980s through early 1990s, caused by the exponentially high cost of creating and maintaining expert digital information databases. Despite waning public interest during the latter period, the pioneers in the field of artificial intelligence continued to collaborate to advance AI. The first AIM workshop funded by the National Institutes of Health was conducted at Rutgers University in 1975. These instances exemplify the pioneers’ early cooperation in AIM.(1)
Terry Allen Winograd: a Stanford University computer science professor well-known in the areas of applied philosophy and artificial intelligence, studied and demonstrated computers’ capacity to comprehend English phrases in a limited environment of children’s blocks. After integrating his language comprehension software, SHRDLU (natural language comprehension computer program), with a robot arm, the latter was able to carry out orders written in English.(2)
Terry Allen Winograd: a Stanford University computer science professor well-known in the areas of applied philosophy and artificial intelligence, studied and demonstrated computers’ capacity to comprehend English phrases in a limited environment of children’s blocks.
After integrating his language comprehension software, SHRDLU (natural language comprehension computer program), with a robot arm, the latter was able to carry out orders written in English.(2)
The Boyer-Moore theorem prover, a piece of software that assists users in solving issues and answering questions that require reasoning, was initially intended to be a completely automated theorem prover for logic. Theorem prover’s key concepts were using the programming language Lisp as a working logic, the dependence on a defining principle for complete recursive functions, substantial rewriting, symbolic evaluation, and an induction heuristic-based approach to the latter.(3)
The Boyer-Moore theorem prover, a piece of software that assists users in solving issues and answering questions that require reasoning, was initially intended to be a completely automated theorem prover for logic. Theorem prover’s key concepts were using the programming language Lisp as a working logic, the dependence on a defining principle for complete recursive functions, substantial rewriting, symbolic evaluation, and an induction heuristic-based approach to the latter.(3)
Alain Colmerauer: a French computer scientist, invented the logic programming language Prolog. He was also a pioneer in artificial intelligence and computational linguistics. Prolog is one of the earliest logic programming languages created and continues to be significantly popular, with both free and commercial versions available. This program coding has been used for theorem proving, expert systems, term rewriting, type systems, automated planning, and natural language processing, where it was originally meant to be employed. Modern Prolog environments enable the development of graphical user interfaces, administrative applications, and networked applications.(4)
Alain Colmerauer: a French computer scientist, invented the logic programming language Prolog. He was also a pioneer in artificial intelligence and computational linguistics. Prolog is one of the earliest logic programming languages created and continues to be significantly popular, with both free and commercial versions available. This program coding has been used for theorem proving, expert systems, term rewriting, type systems, automated planning, and natural language processing, where it was originally meant to be employed.
Modern Prolog environments enable the development of graphical user interfaces, administrative applications, and networked applications.(4)
Earl Sacerdoti created ABSTRIPS, one of the earliest hierarchical planning systems. ABSTRIPS generated an ordering of domain literals per condition semi-automatically, established abstraction layers for planning, identified solutions at the most abstract level, and refined them iteratively down to the ground level. If it cannot resolve a subproblem inside an assignment, it reverts to the previous higher level.(5)
Freddy Robot was developed by the Assembly Robotics Group at the University of Edinburgh. Freddy was one of the first robots to include vision, manipulation, intelligent systems, and software flexibility to facilitate retraining/reprogramming for new jobs. The system programmed the arm motions using an innovative set of high-level procedures; the system could reuse these procedures for each new assignment. The most notable accomplishment was the Versatile Assembly Program, which allowed the robot to assemble a toy boat and a toy vehicle from a jumble of mismatched components dumped onto the table.(6)
Freddy Robot was developed by the Assembly Robotics Group at the University of Edinburgh. Freddy was one of the first robots to include vision, manipulation, intelligent systems, and software flexibility to facilitate retraining/reprogramming for new jobs. The system programmed the arm motions using an innovative set of high-level procedures; the system could reuse these procedures for each new assignment.
The most notable accomplishment was the Versatile Assembly Program, which allowed the robot to assemble a toy boat and a toy vehicle from a jumble of mismatched components dumped onto the table.(6)
References