University of Illinois

International Business Machines Corporation

Bo Li

Jinjun Xiong

Yada Zhu

Pin-Yu Chen

Xiuling Li

Kumar Bhaskaran

Jean-Pierre Leburton

Olgica Milenkovic

Charles M Schroeder

SeyedKasra Tabatabaei

Apratim Khandelwal

Nagendra B Athreya

Yunan Ye

A processor training a reinforcement learning model can include receiving a first dataset representing an observable state in reinforcement learning to train a machine to perform an action. The processor receives a second dataset. Using the second dataset, the processor trains a machine learning classifier to make a prediction about an entity related to the action. The processor extracts an embedding from the trained machine learning classifier, and augments the observable state with the embedding to create an augmented state. Based on the augmented state, the processor trains a reinforcement learning model to learn a policy for performing the action, the policy including a mapping from state space to action space.

State-augmented reinforcement learning

The present disclosure provides systems and methods that can provide portable, real-time accessible DNA memories. An example DNA-based data storage system includes a loading region configured to receive a plurality of DNA-based data storage elements in a suspension fluid and a plurality of microtubes disposed in a capture/release region. The microtubes are configured to capture and release the DNA-based data storage elements. The DNA-based data storage system also includes a linearization region configured to linearize the DNA-based data storage elements and a readout region with a readout device configured to provide information indicative of the respective DNA-based data storage elements.

On-chip nanoscale storage system using chimeric DNA

Growing community of inventors

Champaign, IL, United States of America

Bo Li