scholarly journals Analysis of Watson's Strategies for Playing Jeopardy!

2013 ◽  
Vol 47 ◽  
pp. 205-251 ◽  
Author(s):  
G. Tesauro ◽  
D. C. Gondek ◽  
J. Lenchner ◽  
J. Fan ◽  
J. M. Prager
Keyword(s):  
Decision Making ◽  
Question Answering ◽  
Strategic Decision ◽  
Future Event ◽  
Performance Improvements ◽  
Heuristic Strategies ◽  
Quantitative Performance ◽  
Ibm Watson ◽  
Game State ◽  
Confidence Estimates

Major advances in Question Answering technology were needed for IBM Watson to play Jeopardy! at championship level -- the show requires rapid-fire answers to challenging natural language questions, broad general knowledge, high precision, and accurate confidence estimates. In addition, Jeopardy! features four types of decision making carrying great strategic importance: (1) Daily Double wagering; (2) Final Jeopardy wagering; (3) selecting the next square when in control of the board; (4) deciding whether to attempt to answer, i.e., "buzz in." Using sophisticated strategies for these decisions, that properly account for the game state and future event probabilities, can significantly boost a player's overall chances to win, when compared with simple "rule of thumb" strategies. This article presents our approach to developing Watson's game-playing strategies, comprising development of a faithful simulation model, and then using learning and Monte-Carlo methods within the simulator to optimize Watson's strategic decision-making. After giving a detailed description of each of our game-strategy algorithms, we then focus in particular on validating the accuracy of the simulator's predictions, and documenting performance improvements using our methods. Quantitative performance benefits are shown with respect to both simple heuristic strategies, and actual human contestant performance in historical episodes. We further extend our analysis of human play to derive a number of valuable and counterintuitive examples illustrating how human contestants may improve their performance on the show.

scholarly journals Dynamic Decision Making and Race Games

2013 ◽  
Vol 2013 ◽  
pp. 1-15
Author(s):  
Shipra De ◽  
Darryl A. Seale
Keyword(s):  
Decision Making ◽  
Survey Data ◽  
Optimal Policy ◽  
Optimal Strategy ◽  
Dynamic Decision Making ◽  
Complex Nature ◽  
Heuristic Strategies ◽  
Game State ◽  
Frequent Criticism ◽  
Decision Tasks

Frequent criticism of dynamic decision making research pertains to the overly complex nature of the decision tasks used in experimentation. To address such concerns, we study dynamic decision making with respect to a simple race game, which has a computable optimal strategy. In this two-player race game, individuals compete to be the first to reach a designated threshold of points. Players alternate rolling a desired quantity of dice. If the number one appears on any of the dice, the player receives no points for his turn; otherwise, the sum of the numbers appearing on the dice is added to the player's score. Results indicate that although players are influenced by the game state when making their decisions, they tend to play too conservatively in comparison to the optimal policy and are influenced by the behavior of their opponents. Improvement in performance was negligible with repeated play. Survey data suggests that this outcome could be due to inadequate time for learning or insufficient player motivation. However, some players approached optimal heuristic strategies, which perform remarkably well.

scholarly journals From Control to Coordination: Mahalanobis Distance – Pattern Approach to Strategic Decision Making

2020 ◽  
Author(s):  
Shuichi Fukuda
Keyword(s):  
Decision Making ◽  
Euclidean Space ◽  
Mahalanobis Distance ◽  
Performance Indicator ◽  
Strategic Decision ◽  
Strategic Decision Making ◽  
Adaptive Network ◽  
Pattern Approach ◽  
Quantitative Performance ◽  
Short Time

Up to now, our engineering has been control-based. The goal was clear, We made efforts to get to the goal faster and more effectively. This was possible because changes were smooth, so we could differentiate them mathematically. Therefore, we could predict the future. But today, changes become sharp and unpredictable. Therefore, adaptability becomes more important than efficiency. Or to describe it another way, we need to find an appropriate goal first by trial and error. Yesterday tactics was important, but today strategy becomes important. But our world becomes so complex and complicated that we need to work as a team. And the formation of this team must vary from situation to situation to win the game. Therefore, we must develop a truly adaptive network. This paper proposes that if we introduce non-Euclidean Mahalanobis Distance (MD) and combine it with patterns, then we can develop a holistic and quantitative performance indicator. It evaluates our performance in a very short time for almost any number of dimensions, because it is free from the constraints of orthonormality and units in Euclidean Space. Thus, this Mahalanobis Distance-Pattern (MDP) Approach helps us to make an approriate strategic decision.

scholarly journals HR Managers’ Emotions in Strategic Decision-Making Events: Evidence from Croatia

2021 ◽  
Vol 13 (2) ◽  
pp. 845
Author(s):  
Marli Gonan Božac ◽  
Katarina Kostelić
Keyword(s):  
Decision Making ◽  
Problem Solving ◽  
Strategic Planning ◽  
Strategic Decision ◽  
Strategic Decision Making ◽  
Personal Perspective ◽  
Human Resource Managers ◽  
Organizational Perspective ◽  
Hr Managers ◽  
Experienced Emotion

The inclusion of emotions in the strategic decision-making research is long overdue. This paper deals with the emotions that human resource managers experience when they participate in a strategic problem-solving event or a strategic planning event. We examine the patterns in the intensity of experienced emotions with regard to event appraisal (from a personal perspective and the organization’s perspective), job satisfaction, and coexistence of emotions. The results reveal that enthusiasm is the most intensely experienced emotion for positively appraised strategic decision-making events, while frustration is the most intensely experienced emotion for negatively appraised problem-solving events, as is disappointment for strategic planning. The distinction between a personal and organizational perspective of the event appraisal reveals differences in experienced emotions, and the intensity of experienced anger is the best indicator of the difference in the event appraisals from the personal and organizational perspective. Both events reveal the variety of involved emotions and the coexistence of—not just various emotions, but also emotions of different dominant valence. The findings indicate that a strategic problem-solving event triggers greater emotional turmoil than a strategic planning event. The paper also discusses theoretical and practical implications.

Beyond Bounded Rationality: CEO Reflective Capacity and Firm Sustainability Performance

2021 ◽  
pp. 1-38
Author(s):  
Yingya Jia ◽  
Anne S. Tsui ◽  
Xiaoyu Yu
Keyword(s):  
Decision Making ◽  
Bounded Rationality ◽  
Three Dimensional ◽  
Strategic Decision ◽  
Future Research ◽  
Sustainability Performance ◽  
Business Executives ◽  
Dimensional Measure ◽  
Reflective Capacity ◽  
Mediating Mechanisms

ABSTRACT Optimal or rational decision making is not possible due to informational constraints and limits in computation capability of humans (March & Simon, 1958; March, 1978). This bounded rationality serves as a filtering process in decision making among business executives (Hambrick & Mason, 1984). In this study, we propose the concept of CEO reflective capacity as a behavior-oriented cognitive capability that may overcome to some extent the pervasive limitation of bounded rationality in executive decision-making. Following Hinkin's (1998) method and two executive samples, we developed and validated a three-dimensional measure of CEO reflective capacity. Based on two-wave surveys of CEOs and their executive-subordinates in 213 Chinese small-medium sized firms, we tested and confirmed three hypotheses on how CEO reflective capacity is related to a firm's sustainability performance (including economic, societal, and environmental dimensions) through the mediating mechanisms of strategic decision comprehensiveness and CEO behavioral complexity. We discuss the contribution of this study to the literature on the upper echelons and information processing perspectives. We also identify the implications for future research on strategic leadership and managerial cognition in complex and dynamic contexts.

Sign in / Sign up

Export Citation Format

Share Document