Sameera Lanka
Papers
1
Total Citations
21
H-Index
1
About
Sameera Lanka is a researcher in artificial intelligence, with a primary focus on deep reinforcement learning (RL) and sample-efficient learning algorithms. Her most notable contribution is the development of ARCHER (Aggressive Rewards to Counter bias in Hindsight Experience Replay), a method that addresses a critical limitation in RL: learning from binary and sparse rewards. By introducing aggressive reward shaping, ARCHER mitigates the bias inherent in standard Hindsight Experience Replay (HER), enabling agents to learn more effectively from limited successful experiences. This work, published in 2018, has garnered 21 citations and is recognized for its practical impact on improving sample efficiency in complex RL tasks. Lanka’s research is particularly valuable for robotic control and autonomous systems, where reward signals are often sparse. Her work exemplifies a deep understanding of algorithmic biases and offers a clear, actionable solution for advancing RL in real-world applications.
Research Focus
Key Achievements
Top Papers
- 1ARCHER: Aggressive Rewards to Counter bias in Hindsight Experience Replay21 citations · 2018