Question

researchers at carnegie mellon university and olin college of engineering have explored the use of contact microphones as an alternative to conventional tactile sensors for training machine learning models in robotic manipulation, hypothesizing that audio data could enhance performance. they pre - trained a self - supervised machine learning model on audio - visual representations from the audioset dataset, which includes over 2 million 10 - second video clips. utilizing audio - visual instance discrimination (avid), the model learned to differentiate between various types of audio - visual data. the researchers concluded that contact microphones provide valuable audio - based information that may make large - scale audio - visual pretraining a worthwhile venture, as they observed that ______
15
mark for review
which choice most logically completes the text?
a despite the sacrifice in overall performance, the inclusion of audio data was able to return certain creative outputs that could not be attained with the more traditional visual - tactile sensors.
b especially when dealing with unfamiliar objects and locations, robotic manipulation became more efficient with the inclusion of audio data when it was processed separately rather than alongside visual data.
c in a series of tests, robots using this model completed real - world manipulation tasks with up to 60 demonstrations per task, outperforming models that relied solely on visual data.
d data from conventional tactile sensors significantly improved performance on real - world manipulation tasks compared to data from only visual sensors that provided monochromatic differentiation for intensity.

Explanation:

The text is about a research on using contact microphones (audio data) for training machine - learning models in robotic manipulation. The researchers' conclusion is that audio - based information from contact microphones makes large - scale audio - visual pretraining worthwhile. We need to find an option that supports this conclusion by showing the positive impact of the model (using audio - visual data) on robotic manipulation.

• Option A: Talks about creative outputs and a sacrifice in overall performance. But the research is about enhancing performance in robotic manipulation, not about creative outputs or sacrificing performance. So, A is incorrect.
• Option B: Says that audio data is more efficient when processed separately from visual data. However, the model in the research is based on audio - visual representations (processing them together), so B goes against the research's approach and is incorrect.
• Option C: Shows that in tests, robots using the model (which uses audio - visual data) outperformed models that used only visual data in real - world manipulation tasks. This supports the idea that audio - visual pretraining (using contact microphones' audio data) is worthwhile as it improves robotic manipulation performance. So, C is correct.
• Option D: Compares conventional tactile sensors and visual sensors, but the research is about using contact microphones (audio data) along with visual data, not about this comparison. So, D is incorrect.

Answer:

C. in a series of tests, robots using this model completed real - world manipulation tasks with up to 60 demonstrations per task, outperforming models that relied solely on visual data.