Why One-step and IQL are imitation-based methods? The core difference between RL-based and imitation-based methods is that RL-based methods learn a value function of policy π while imitation-based methods don't. Learning the value function of π requires off-policy evaluation of π (i.e., learning Qπ or Vπ), which is prone to distribution shift. The policy evaluation and policy improvement will also affect each other as they are coupled. Imitation-based methods don't learn Qπ or Vπ, but some of them do learn a value function.

Neural Information Processing Systems http://nips.cc/

Language encoding models help explain language processing in the human brain by learning functions that predict brain responses from the language stimuli that elicited them.

Identifying the features learned by neural networks is a core challenge in mechanistic interpretability. Sparse autoencoders (SAEs), which learn a sparse, overcomplete dictionary that reconstructs a network's internal activations, have been used to identify these features. However, SAEs may learn more about the structure of the dataset than the computational structure of the network. There is therefore only indirect reason to believe that the directions found in these dictionaries are functionally important to the network. We propose end-to-end (e2e) sparse dictionary learning, a method for training SAEs that ensures the features learned are functionally important by minimizing the KL divergence between the output distributions of the original model and the model with SAE activations inserted.

Identifying the features learned by neural networks is a core challenge in mechanistic interpretability. Sparse autoencoders (SAEs), which learn a sparse, overcomplete dictionary that reconstructs a network's internal activations, have been used to identify these features. However, SAEs may learn more about the structure of the dataset than the computational structure of the network. There is therefore only indirect reason to believe that the directions found in these dictionaries are functionally important to the network. We propose end-to-end (e2e) sparse dictionary learning, a method for training SAEs that ensures the features learned are functionally important by minimizing the KL divergence between the output distributions of the original model and the model with SAE activations inserted.