Search papers, labs, and topics across Lattice.
Affiliation:
2
0
2
8
This work reduces state-adversarial Markov decision processes to a strategically equivalent constrained zero-sum one-sided partially observable stochastic game and presents the first algorithmic route to computing $\epsilon$-approximations of initial-state dependent equilibria.
Disentangled representations can reduce unlearning collateral damage by up to 4x, challenging the assumption that entanglement is merely a theoretical concern.
