|
Projects for Students (suggested and biased by Eyal Amir) Topics include (but not restricted to): Reasoning, Knowledge Representation, Dynamic Systems, Decision Making, Optimization Algorithms |
| Inference Projects |
| Number | Topic | Student |
|---|---|---|
| |
Extension of lock resolution using craig's interpolation theorem | |
| |
Hybrid reasoning with logic and probabilities via partitioning (requires knowledge of FOPL, at least Halpern's work) | |
| |
Logical filtering with a first-order language | |
| |
Logical filtering with BDDs | |
| |
Activity detection using filtering (any method) | |
| |
SLAM2.0 improved and applied to a robotic arm | |
| |
Combining logical filtering and stochastic filtering | |
| |
Survey propagation for dynamic settings (e.g., planning via SAT) | |
| |
Approximation algorithm for hyper-treewidth | |
| |
Implementation of an algorithm for planar treewidth | |
| |
TAKEN (replaced soon) | |
| |
Labeling image segments with words (using a probabilistic graphical model) | |
| |
Implementation of Message-Passing as a restriction strategy for reasoning in FOL | |
| |
Playing Kriegspiel | |
| |
Equational reasoning in first-order probabilistic models | |
| |
First-Order DPLL with equality | |
| |
LSA-like robot control architecture with probabilities | |
| |
Learning action models via filtering | |
| |
MCMC for image segmentation | |
| |
Robot localization for a basketball game | |
| |
Control of a robotic arm via dynamic programming | |
| |
TAKEN (replaced soon) | |
| |
TAKEN (replaced soon) |   |
| |
Orienting polygonal parts using filtering | |
| |
Resolution on reconfigurable hardware |
| Decision-Making Projects |
| Number | Topic | Presenter |
|---|---|---|
| |
Kriegspiel game player | |
| |
First-Order POMDPs | |
| |
Exact Factored MDPs | |
| |
Conformant planning using Logical Filtering | |
| |
Activity detection using filtering (any method) | |
| |
SLAM2.0 improved and applied to a mobile robot | |
| |
Poker Playing | |
| |
Bridge Player | |
| |
First-Order Factored Planning | |
| |
Adventure-Game Exploration using Commonsense knowledge | |
| |
Approximate Deterministic POMDPs via Logical Filtering | |
| |
Probabilistic Resolution in Dynamic Bayesian Networks | |
| |
POMDPs approximated via DBNs | |
| |
Robot control using a POMDP | |
| |
First-Order Reinforcement Learning | |
| |
Reshaping rewards using commonsense knowledge | |
| |
Partially observable LSA-based robot control architecture | |
| |
Reinforcement learning in deterministic domains | |
| |
Controlling a Levitating Robot | |
| |
Robot localization for a basketball game | |
| |
LSA-based control system for a robotic arm |
| Viewing PostScript and PDF |
Depending on the computer you are using, you may be able to download a PostScript viewer or PDF viewer for it if you don't already have one.
| Comments to Eyal Amir |