I have a sort of "new" configuration in the hardware implementing of an SNN or ANN, which i'm going to detail here. I don't know if it has any worth or not, but it's just a concept inside my mind:
We are trying to mimic a neuron cell. Like many other types of cells, we have two pumping mechanisms, Na and K which balance the cell voltage and are behind its spiking function. The most similar design to that concept (as I far as I know) is 2T2R configuration which (I guess mostly in the case of PCMs are useful as they have different setting and resetting behaviors) gives us a better controllability over our nodes. Now, I have another suggestion for this configuration: instead of having an R—T--|--T—R configuration, consider an R—T as our main node’s output and then connect another R—T’s output to its back-gate (connecting either drain or source to the main R-T). Fabrication of this design is harder than the previous design but here, we may achieve something that I may call it the learning rate. In each cycle, the effect of the reducing pump is different from its previous state (hence, as we reach toward the answer, we reduce or increase the effect of back propagation). This design may (or may not) has a drawback which is changing the state of our main FET (its depletion region) but I don’t think it become very considerable. What is your opinion? Does it have any point at all?
Dear Alireza Alikhani ,
Your idea is interesting, however., when no text/experiments/code/manuscript is available it is hard for your nomenclature or research item to be commented on. The best idea is to understand the research methodology and follow it up to peer review publication procedure and get comments on such theories. I just mentioning on how to put forward your such research pathway up to the goals:
Suppose you are perusing a PhD or any other scientific project, the tips may be:
- It may be unsolved (or partially solved, but may need further upgrades or improvements…may be open from decades to solve gradually, e.g., Human GENOME project etc., … ) topic or problem from academia, industry or other end users.
- Crucial to know about the scope and gape of the research topic with getting questions on hands and the answers to be investigated.
- In particular, it is important for the researcher/investigator to know about the importance and benefits of the underline research with goals to convince the academia, industry or other end users for getting research responsibility and grants.
- To make literature review by reading and understanding most of the previous research findings related to the current research.
- To evaluate the exact gape and then to gain further theoretical and computational knowledge with tools, related to the filling/solving of research gape or upgrade of the research questions at hand.
- To outline the research steps and enhanced understanding and improvements in its development month by month within the given time-frame of the research accomplishment.
- To accomplish and meet all the research goals (of the PhD/Project) and in some or many occasions to share it with the global scientific community by presentations to audiences and publications in conferences and journals.
- The phase of research presenting to the live audiences and its publishing in conference and journals provides golden opportunities to get valuable discussions and comments from the other research experts.
- By passing through the above phases, the PhD candidate or project solver has to write a complete report of their corresponding PhD research work (THESIS) or project report.
- The report must have significant research impact that may be reviewed and acknowledged by some expert referees within the related research area.
- At this stage the outcomes may be in the form of accepts, accepts with minor review, accept with major review, acceptable for the Master degree only, or total reject.
Please read my paper
An Adaptive Filter to Pick up a Wiener Filter from the Error using MSE with and Without Noise
I hope you find a good answer.
- Badges
- Science topic