Yes, I do not know how the use of latent actions helps when they actually deploy. Either they are completely different mode not used at all,or they have negative influence on the whole policy since these actions are not GT and are much coarser level, and obviously the frames can be unfaithful.
March 19, 2025 at 2:09 PM
Yes, I do not know how the use of latent actions helps when they actually deploy. Either they are completely different mode not used at all,or they have negative influence on the whole policy since these actions are not GT and are much coarser level, and obviously the frames can be unfaithful.
Beautiful paper, and beautiful visaluzations!
March 12, 2025 at 6:21 PM
Beautiful paper, and beautiful visaluzations!
Omg, thank you so much!
March 10, 2025 at 7:42 PM
Omg, thank you so much!
Yes, I agree. I am thinking of a distinct future when we have all the resources in our solar system. Lol, too hypothetical!
March 2, 2025 at 9:57 PM
Yes, I agree. I am thinking of a distinct future when we have all the resources in our solar system. Lol, too hypothetical!
Fine-tuning is a fine game when you know the test-distribution and I believe that active fine-tuning at test time will be on the cards. But none can truly generalize out-of-distribution. You either collect a hell lot of data or stumble upon an architecture that truly generalizes of OOD for your task
March 2, 2025 at 9:55 PM
Fine-tuning is a fine game when you know the test-distribution and I believe that active fine-tuning at test time will be on the cards. But none can truly generalize out-of-distribution. You either collect a hell lot of data or stumble upon an architecture that truly generalizes of OOD for your task
Example the bias of CNNs to exploit high-frequency components in image might be good for CV benchmark performance, if you feed the same representations to a robotic policy then it is clearly sub-optimal, shape is more important here ....
March 2, 2025 at 9:54 PM
Example the bias of CNNs to exploit high-frequency components in image might be good for CV benchmark performance, if you feed the same representations to a robotic policy then it is clearly sub-optimal, shape is more important here ....
Well, is it efficient, yes! But is it the optimal, probably no. The problem with pre-trained modules is that their representations are biased by their objective to fit a benchmark. And this objective encapsulates years of community based benchmark overfitting that we don't even think of biases ..
March 2, 2025 at 9:53 PM
Well, is it efficient, yes! But is it the optimal, probably no. The problem with pre-trained modules is that their representations are biased by their objective to fit a benchmark. And this objective encapsulates years of community based benchmark overfitting that we don't even think of biases ..
I think pure end-to-end learning will be a deployment scenario. Most of the model based planning methods and priors will be used for data generation in simulation for efficient exploration in data collection. But it is hard to win over pure end-to-end learning system if you have the data once.
March 2, 2025 at 7:22 PM
I think pure end-to-end learning will be a deployment scenario. Most of the model based planning methods and priors will be used for data generation in simulation for efficient exploration in data collection. But it is hard to win over pure end-to-end learning system if you have the data once.
Yes, compared to previous years it has become much more accessible and more like typical supervised learning, which I believe is great. Also looking forward towards your lab opening and following your research. Daniel Brown talks about you!
November 18, 2024 at 8:45 PM
Yes, compared to previous years it has become much more accessible and more like typical supervised learning, which I believe is great. Also looking forward towards your lab opening and following your research. Daniel Brown talks about you!
Interesting,I think most people don't apply robotics as an application area because of slow feedback loop. I also don't think that architectures developed for other applications may not be the most optimal for robotics. I believe it depends on the assumption you are making on the data distribution.
November 18, 2024 at 5:58 PM
Interesting,I think most people don't apply robotics as an application area because of slow feedback loop. I also don't think that architectures developed for other applications may not be the most optimal for robotics. I believe it depends on the assumption you are making on the data distribution.
Hmm, interesting. What do you think is the use case of RAG in robotics? Like Behavior Retrieval?(arxiv.org/abs/2304.08742)
November 17, 2024 at 7:21 PM
Hmm, interesting. What do you think is the use case of RAG in robotics? Like Behavior Retrieval?(arxiv.org/abs/2304.08742)