Mastering Uncertainty: An Effective Approach to Training Machines for Real-World Situations — Researchers develop an alg |
MIT and Technion researchers have developed an adaptive algorithm that optimizes machine learning by combining imitation and reinforcement learning. The algorithm autonomously decides when to follow or diverge from a teacher model, improving training efficiency and effectiveness. This approach presents a potential way to improve training for complex tasks and could potentially be used with larger models like GPT-4 to train smaller, task-focused models.
Their dynamic approach allows the student to diverge from copying the teacher when the teacher is either too good or not good enough, but then return to following the teacher at a later point in the training process if doing so would achieve better results and faster learning. “This combination of learning by trial-and-error and following a teacher is very powerful.
By dynamically determining which method achieves better results, the algorithm is adaptive and can pick the best technique throughout the training process. Thanks to this innovation, it is able to more effectively teach students than other methods that aren’t adaptive, Shenfeld says. Reorienting objects is one among many manipulation tasks that a future home robot would need to perform, a vision that the Improbable AI lab is working toward, Agrawal adds.
Nigeria Latest News, Nigeria Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
Seniors concerned about Social Security amid debt limit uncertaintyAmid political back-and-forth over the debt limit, Scripps News spoke with retirees who are concerned about the future of their Social Security check.
Read more »
Oil falls on US debt deal struggles, OPEC+ talks uncertaintyOil fell on Tuesday, giving up earlier gains, as concerns about the viability of the U.S. debt ceiling pact cooled the market's risk-on sentiment and mixed messages from major producers have clouded the supply outlook ahead of their meeting this weekend.
Read more »
Stock-market uncertainty could rise after debt-ceiling deal: Morgan StanleyInsider tells the global tech, finance, markets, media, healthcare, and strategy stories you want to know.
Read more »