There is an advanced algorithm created by Toshiba Corporation that improves the effectiveness of its Simulated Bifurcation Machine. This machine is a proprietary quantum computer, specifically engineered to solve combinatorial optimization problems. The algorithm greatly improves the likelihood of achieving optimal solutions through trial runs. Experts refer to this measure as success probability, which remains a key benchmark for evaluating optimization technologies. The SBM supports large-scale problem solving in fields like drug discovery, route optimization, and investment portfolio design.
Earlier algorithms required many trials to find optimal solutions, especially when solving large-scale optimization challenges. However, these problems often trap the search process in local optima, reducing efficiency under limited trial conditions. To combat the challenge above, Toshiba developed a third-generation simulated bifurcation algorithm to enhance computing performance. This development was done after the launch of the firm’s first algorithm in April 2019 and its second algorithm, which was introduced in February 2021. Both algorithms performed satisfactorily when solving various optimization problems.
Edge of Chaos Unlocks Faster Optimization
The novel algorithm employs separate parameters of bifurcations rather than a universal parameter of control used previously. Every bifurcation parameter becomes linked directly to a particular value of the position variable. The algorithm separately regulates each bifurcation parameter depending on the corresponding position variable’s value. This makes it possible for the algorithm to find an optimal solution in a very efficient manner, regardless of the complexity or size of the problem under consideration. Thus, the algorithm behaves in a stable or chaotic manner.
Toshiba discovered that controlling behavior at the edge of chaos improves optimization efficiency significantly. This is because it is the critical juncture where the behavior becomes unpredictable from deterministic behaviors. In utilizing this border, the algorithm avoids being trapped in local optima. In doing so, the likelihood of success increases towards near-perfect optimization results. Compared to its previous generation, the advanced SBM reaches its results up to 100 times quicker than before. With less computation time involved, this leads to a faster deployment in practical optimization problems. Such innovations will hasten progress within sectors like healthcare and finance.
Explore IT Tech News for the latest advancements in Information Technology & insightful updates from industry experts!
News Source: Businesswire.com
