Rushiv Arora
@rushivarora.bsky.social
Special thanks to @eugenevinitsky.bsky.social for being an amazing mentor!
October 8, 2025 at 3:19 PM
Special thanks to @eugenevinitsky.bsky.social for being an amazing mentor!
2. LEXPOL’s performance benchmarks against previous methods on MetaWorld
3. A combination of LEXPOL with the previous natural language based state embedding algorithm, giving a joint method combing state and action factorization
3. A combination of LEXPOL with the previous natural language based state embedding algorithm, giving a joint method combing state and action factorization
October 8, 2025 at 3:19 PM
2. LEXPOL’s performance benchmarks against previous methods on MetaWorld
3. A combination of LEXPOL with the previous natural language based state embedding algorithm, giving a joint method combing state and action factorization
3. A combination of LEXPOL with the previous natural language based state embedding algorithm, giving a joint method combing state and action factorization
Paper Highlights:
1. Qualitative analysis of LEXPOL (end-to-end learning) and frozen pre-trained single-task policies. We note that LEXPOL successfully disentangles the tasks into fundamental skills, and learns to combine them without a decomposition to primitive actions.
1. Qualitative analysis of LEXPOL (end-to-end learning) and frozen pre-trained single-task policies. We note that LEXPOL successfully disentangles the tasks into fundamental skills, and learns to combine them without a decomposition to primitive actions.
October 8, 2025 at 3:19 PM
Paper Highlights:
1. Qualitative analysis of LEXPOL (end-to-end learning) and frozen pre-trained single-task policies. We note that LEXPOL successfully disentangles the tasks into fundamental skills, and learns to combine them without a decomposition to primitive actions.
1. Qualitative analysis of LEXPOL (end-to-end learning) and frozen pre-trained single-task policies. We note that LEXPOL successfully disentangles the tasks into fundamental skills, and learns to combine them without a decomposition to primitive actions.
LEXPOL is inspired by our ability to combine multiple different sub-skills together to solve larger tasks based on context. It works by factorizing the complexity of multi-task reinforcement learning into smaller learnable pieces.
October 8, 2025 at 3:19 PM
LEXPOL is inspired by our ability to combine multiple different sub-skills together to solve larger tasks based on context. It works by factorizing the complexity of multi-task reinforcement learning into smaller learnable pieces.
This helps since multi-task RL is hard because a single monolithic policy must entangle many skills. LEXPOL factorizes control into smaller learnable pieces and uses language as the router for composition.
October 8, 2025 at 3:19 PM
This helps since multi-task RL is hard because a single monolithic policy must entangle many skills. LEXPOL factorizes control into smaller learnable pieces and uses language as the router for composition.
The idea: give the agent a natural-language task description (“push the green button”) and let a learned language gate blend or select among several sub-policies (skills) based on context. One shared state; multiple policies; a gating MLP guided by language embeddings chooses the action.
October 8, 2025 at 3:19 PM
The idea: give the agent a natural-language task description (“push the green button”) and let a learned language gate blend or select among several sub-policies (skills) based on context. One shared state; multiple policies; a gating MLP guided by language embeddings chooses the action.
This is a cause that is very close to my heart, and I am glad to have found the NOCC. They do a lot of important work for patients and their caregivers.
February 12, 2025 at 3:21 AM
This is a cause that is very close to my heart, and I am glad to have found the NOCC. They do a lot of important work for patients and their caregivers.
My grandmother passed away from ovarian cancer in 1995. I never got the chance to meet her, but I have always felt extremely connected with her through the countless stories and family heirlooms my parents have shared with me.
February 12, 2025 at 3:21 AM
My grandmother passed away from ovarian cancer in 1995. I never got the chance to meet her, but I have always felt extremely connected with her through the countless stories and family heirlooms my parents have shared with me.
This paper is personally meaningful to me since it is my first solo-authored paper! (And 7th overall!).
I got the idea in 2023 in the final months of my master’s but didn’t have the chance to work on it until last year. I am thrilled to finally publish it!
January 13, 2025 at 4:04 PM
This paper is personally meaningful to me since it is my first solo-authored paper! (And 7th overall!).
I got the idea in 2023 in the final months of my master’s but didn’t have the chance to work on it until last year. I am thrilled to finally publish it!
Read the paper to explore how H-UVFAs advance scalable and reusable skills in RL! #ReinforcementLearning #MachineLearning #AI
January 13, 2025 at 4:04 PM
Read the paper to explore how H-UVFAs advance scalable and reusable skills in RL! #ReinforcementLearning #MachineLearning #AI
- Outperforming UVFAs: In Hierarchical settings, H-UVFAs have superior performance and generalization than UVFAs. In fact, UVFAs failed to learn in some settings.
- Learning in both supervised and reinforcement learning contexts.
- Learning in both supervised and reinforcement learning contexts.
January 13, 2025 at 4:04 PM
- Outperforming UVFAs: In Hierarchical settings, H-UVFAs have superior performance and generalization than UVFAs. In fact, UVFAs failed to learn in some settings.
- Learning in both supervised and reinforcement learning contexts.
- Learning in both supervised and reinforcement learning contexts.
Core Contributions:
- Hierarchical Embeddings: We show that it is possible to break down hierarchical value functions into its core elements by leveraging higher-order decomposition methods in Mathematics like Tucker Decompositions.
- Zero-shot generalization: H-UVFAs can extrapolate to new goals!
- Hierarchical Embeddings: We show that it is possible to break down hierarchical value functions into its core elements by leveraging higher-order decomposition methods in Mathematics like Tucker Decompositions.
- Zero-shot generalization: H-UVFAs can extrapolate to new goals!
January 13, 2025 at 4:04 PM
Core Contributions:
- Hierarchical Embeddings: We show that it is possible to break down hierarchical value functions into its core elements by leveraging higher-order decomposition methods in Mathematics like Tucker Decompositions.
- Zero-shot generalization: H-UVFAs can extrapolate to new goals!
- Hierarchical Embeddings: We show that it is possible to break down hierarchical value functions into its core elements by leveraging higher-order decomposition methods in Mathematics like Tucker Decompositions.
- Zero-shot generalization: H-UVFAs can extrapolate to new goals!
We extend Universal Value Function Approximators (UVFAs) to hierarchical RL, enabling zero-shot generalization across new goals in multi-task settings while retaining the benefits of temporal abstraction.
January 13, 2025 at 4:04 PM
We extend Universal Value Function Approximators (UVFAs) to hierarchical RL, enabling zero-shot generalization across new goals in multi-task settings while retaining the benefits of temporal abstraction.