COMPASS (Criticality, Options & Metrics for Plan Analysis via Stochastic Simulation)
The main goal is to help human planners discover the what, where and why of risk and inefficiency. Uncertainty, in its many forms, is the source of deviation of a plan from its intended course, making it possible to unfold in a multitude of ways. The key to making better plans is understanding what the potential executions might be, where in those execution paths the problems are, and why they occurred.
The main goal is to help human planners discover the what, where and why of risk and inefficiency. Uncertainty, in its many forms, is the source of deviation of a plan from its intended course, making it possible to unfold in a multitude of ways. The key to making better plans is understanding what the potential executions might be, where in those execution paths the problems are, and why they occurred.
REGES (Rapidly-formed Electrical Grids for Extreme Scenarios)
"Typically, electrical grids are designed, operated and managed from the top down in a hierarchical spoke architecture. The proposed effort reverses this approach, and proposes to research the establishment of energy distribution architectures that are decentralized and ad-hoc, for domains where they cannot be reliably pre-planned, and where management is inherently decentralized and must be automated."
"Typically, electrical grids are designed, operated and managed from the top down in a hierarchical spoke architecture. The proposed effort reverses this approach, and proposes to research the establishment of energy distribution architectures that are decentralized and ad-hoc, for domains where they cannot be reliably pre-planned, and where management is inherently decentralized and must be automated."
COSMOS (COmputational Simulation and Modeling Of Societies)
"The COSMOS project studies the modeling and simulation of human social behavior. The work undertaken spans basic research in the computational modeling of human social systems as well as the application of these models to a range of gaming applications, including simulation-based serious games and interactive narratives."
"The COSMOS project studies the modeling and simulation of human social behavior. The work undertaken spans basic research in the computational modeling of human social systems as well as the application of these models to a range of gaming applications, including simulation-based serious games and interactive narratives."
MICE (Modeling Interaction in Cooperative-competitive Environments)
This project seeks to create data-driven, population-specific models of human behavior by learning large amounts of detailed game data collected from multi-player online games, such as the Social Ultimatum Game and TeamIt. We combine statistical machine learning methods with these data sources to create well-grounded, belief-based probabilistic human behavior models.
This project seeks to create data-driven, population-specific models of human behavior by learning large amounts of detailed game data collected from multi-player online games, such as the Social Ultimatum Game and TeamIt. We combine statistical machine learning methods with these data sources to create well-grounded, belief-based probabilistic human behavior models.
CSC (Criticality-Sensitive Coordination)
"The goal is creating distributed intelligent software systems that will help fielded units adapt their mission plans as the situation around them changes and impacts their plans. Intelligent software Coordinators do this by reasoning about the tasks assigned to a given unit, the task timings, how the tasks interact with those of other units, and by evaluating possible changes such as changing task timings, task assignments, or selecting from pre-planned contingencies."
"The goal is creating distributed intelligent software systems that will help fielded units adapt their mission plans as the situation around them changes and impacts their plans. Intelligent software Coordinators do this by reasoning about the tasks assigned to a given unit, the task timings, how the tasks interact with those of other units, and by evaluating possible changes such as changing task timings, task assignments, or selecting from pre-planned contingencies."
C-MARCH (Coordinated MANETs using RF-Guided Criticality-focused Hormones)
"This program will develop intelligent autonomous radio relay nodes that exploit movement to establish and manage mesh networks in urban settings. The goal is to create small, inexpensive, smart robotic radio relay nodes that dismounted warfighters drop as they deploy in urban settings. The nodes then self-configure and form a mesh network - a temporary infrastructure that establishes communications over the region.
"This program will develop intelligent autonomous radio relay nodes that exploit movement to establish and manage mesh networks in urban settings. The goal is to create small, inexpensive, smart robotic radio relay nodes that dismounted warfighters drop as they deploy in urban settings. The nodes then self-configure and form a mesh network - a temporary infrastructure that establishes communications over the region.
CALO (Collaborative Agent that Learns and Organizes)
"The mission is to radically improve the way computers support humans by enabling systems that are cognitive, i.e., computer systems that can reason, learn from experience, be told what to do, explain what they are doing, reflect on their experience, and respond robustly to surprise. More specifically, CALO will develop a series of prototype cognitive systems that can act as an assistant for commanders and staff. "
"The mission is to radically improve the way computers support humans by enabling systems that are cognitive, i.e., computer systems that can reason, learn from experience, be told what to do, explain what they are doing, reflect on their experience, and respond robustly to surprise. More specifically, CALO will develop a series of prototype cognitive systems that can act as an assistant for commanders and staff. "
ActComm Project on Transportable Agents in Wireless Networks
"The goal of the ActComm project is to develop technologies that will maximize the usability of complex, global computer and communications networks, focusing especially on wireless networks for modern command-and-control applications. The main technical innovation is the concept of an active communications system consisting of dynamic elements - active software, active information, active hybrid networks and active resource allocation."
"The goal of the ActComm project is to develop technologies that will maximize the usability of complex, global computer and communications networks, focusing especially on wireless networks for modern command-and-control applications. The main technical innovation is the concept of an active communications system consisting of dynamic elements - active software, active information, active hybrid networks and active resource allocation."
BLEMS (Building-Level Energy Management Systems)
We seek to reduce energy consumption and increase occupancy comfort by combining behavioral models of occupants with optimal decision-making on energy purchasing and allocation. Our live test-bed will be deployed on the USC campus, where the developed algorithms will automatically learn patterns of building use and occupant behaviors, and adjust operations to optimize energy use.
We seek to reduce energy consumption and increase occupancy comfort by combining behavioral models of occupants with optimal decision-making on energy purchasing and allocation. Our live test-bed will be deployed on the USC campus, where the developed algorithms will automatically learn patterns of building use and occupant behaviors, and adjust operations to optimize energy use.
CADS (Cancer As a Dynamic Stochastic-graphical game)
We model cancer cells as selfish autonomous agents engaging in a dynamic stochastic graphical game. This multi-agent view addresses important properties such as heterogeneous behavior, probabilistic outcomes and a changing environment. We seek insights and tools to better generate, understand and investigate more complex phenomena in cancer cell populations.
We model cancer cells as selfish autonomous agents engaging in a dynamic stochastic graphical game. This multi-agent view addresses important properties such as heterogeneous behavior, probabilistic outcomes and a changing environment. We seek insights and tools to better generate, understand and investigate more complex phenomena in cancer cell populations.
GAMBIT (Geospatial Analysis of Motion-Based Intelligence and Tracking)
Movement data can be combined with geospatial information and transformed into probabilistic graphical models that represent both social and temporal relationships between objects in the observed area. We then apply machine-learning techniques to cluster patterns in these graphical models to assist human users in performing strategic level analysis such as behavior prediction and anomaly detection.
Movement data can be combined with geospatial information and transformed into probabilistic graphical models that represent both social and temporal relationships between objects in the observed area. We then apply machine-learning techniques to cluster patterns in these graphical models to assist human users in performing strategic level analysis such as behavior prediction and anomaly detection.
TeamIt + TrackIt
We investigate collaboration and coordination among multiple teams and individuals, which arises in many domains including disaster response. TrackIt enables real-time situational awareness and task coordination the TeamIt Alternate Reality Game (ARG) enables team-based human experiments that run on iPads/iPhones and test real-world scenarios in the field.
We investigate collaboration and coordination among multiple teams and individuals, which arises in many domains including disaster response. TrackIt enables real-time situational awareness and task coordination the TeamIt Alternate Reality Game (ARG) enables team-based human experiments that run on iPads/iPhones and test real-world scenarios in the field.
Spatiotemporal Patterns in Basketball
New technologies deployed in the NBA enable automatic collection of fine-grained spatio-temporal data that tracks player and ball movements, opening up a wide range of pattern recognition and sports analysis problems. We develop and apply machine learning, data analysis, and visualization tools to both validate long-held beliefs and uncover the hidden truths behind the game.
New technologies deployed in the NBA enable automatic collection of fine-grained spatio-temporal data that tracks player and ball movements, opening up a wide range of pattern recognition and sports analysis problems. We develop and apply machine learning, data analysis, and visualization tools to both validate long-held beliefs and uncover the hidden truths behind the game.
Starcraft AI
Real-time strategy games have been a popular genre for decades, and Starcraft remains one of its most popular exemplars. We are developing a bot to compete in the annual Starcraft AI competition, both in the hopes of galactic domination and as a platform for AI education and research.
Real-time strategy games have been a popular genre for decades, and Starcraft remains one of its most popular exemplars. We are developing a bot to compete in the annual Starcraft AI competition, both in the hopes of galactic domination and as a platform for AI education and research.
