Reducing Impact of CNFET Process Imperfections on Shape of Activation Function by Using Connection Pruning and Approximate Neuron Circuit
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Deep Neural Networks (DNNs) based on Carbon nanotube field effect transistor (CNFET) technology can leverage the potential energy benefits of CNFET based technology in comparison to conventional Si technology. However, like other emerging materials based technologies, the current fabrication processes for CNFETs lack the quality, resulting in CNFETs suffering from process imperfections, consequently degradation in circuit-level performance. Such imperfections will cause timing failure and distort the shape of non-linear activation functions, which are vital in DNN, leading to significant degradation in classification accuracy. We utilize pruning of synaptic weights which combined with proposed approximate neuron circuit significantly reduces the chance of timing failure, and achieve better frequency of operation (speed), even using highly imperfect process. In our example, the proposed configuration with approximate neuron and pruning at a high imperfect process <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(PCNT_{open}=\ 40\%)$</tex> , in comparison to base configuration of precise neuron and no pruning with ideal process <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(PCNT_{open}=\ 0\%)$</tex> , achieves peak accuracy only 0.19% less, but significant energy-delay-product (EDP) advantage (56.7% less), at no area penalty.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it