Nitish Dashora

@DashoraNitish

PhD student @MIT in ML + robot learning | UC Berkeley EECS w/ Honors | Prev @nasa, @amazon, @berkeley_ai | Astronaut+Goldwater Scholar, NSF GRFP

nitishdashora.com

Joined April 2019

516 Following

1,110 Followers

Replies Reposts Media

Nitish Dashora
@DashoraNitish
Oct 20
Crazy dexterity and high speeds unlocked through a VFA
Pulkit Agrawal
@pulkitology
Oct 20
@EkaRobotics has raised a $150M Series A. Generality v/s Performance is the fundamental tension at the frontier of robotics. Generality without performance is just a prototype. Performance without generality is just a single-purpose tool. Customers shouldn’t have to choose. Eka
1
Nitish Dashora
@DashoraNitish
Oct 19
Training for blind manipulation pressures your policy to learn an intelligent "searching" mechanism in context!
Antonia Bronars
@BronarsToni
Oct 19
Every robot already carries contact sensors: its own joints. When a robot touches something, its joints drift slightly off their commands, and the encoders pick it up. Legged robots and humanoids already leverage this signal to walk blind - but is it enough for blind precise
Nitish Dashora
@DashoraNitish
Oct 16
The humanoid sim2real goat dropped another banger, check it out!
Nolan Fey
@nolan_fey
Oct 16
(1/9) Same rewards. Less domain randomization. Higher jumps. Stable landings. Just by improving the actuator model. Introducing Contact-UAN: a residual world model for humanoids, learned from proprioception. No torque sensing or motion capture. Project website:
Nitish Dashora
@DashoraNitish
Oct 14
Crazy stuff: SGS + OmniReset on multi-task RL!
Octi Zhang
@octi_zhang
Oct 13
(1/6) From @GoogleDeepMind’s AlphaGo to AlphaZero, we saw machines find moves that reached beyond what we’d imagined for them. No one taught them those strategies. They emerged from reinforcement learning: playing, failing, and trying again under a single binary task reward for
1
Nitish Dashora
@DashoraNitish
Oct 14
Nice idea to integrate memory supervision with policy training!
Yilin Wu ✈️ CoRL 2026
@YilinWu11
Oct 14
What to remember in the past to act—spatial, temporal or motion details? We often assume what’s useful, but that’s hard to know in advance. Meet IMPRINT 🧠: retain rich memory first, then learn to use it for control. Compact latents keep inference fast as history grows. 🧵 1/10

Nitish Dashora

Nitish Dashora’s posts