# 2020/10/27 meeting

---

1. Past work
2. probability of decision exchange
3. Inverse spikes intervel

---

## past work

---

1. Decision make;corr of recurrent & noise

要在關閉input後保留current like this
放大decision部份後長這樣，約300units -> 1/5sec

if noise upup -> recurrent have to upup too
如此才能維持那一小段decision time

---

2. probability of Decision exchange

### process
| time | neuron | inputstr |
| ---- | ------ | -------- |
| 3000 | 0 | 6 |
| 3000 | 1 | 5 |
| next | next | next |
| var | 0 | 0 |
| var | 1 | 5 |
| next | next | next |
| 3000 | 0 | 6 |
| 3000 | 1 | 5 |

---

3. inverse spikes interval


    Epu = (external_current / -_g) + _rest;
    Z = (_threshold - _reset) / (Epu - _threshold);
    ISI = (1 / ((1 / -_g) * log(1 + Z)));

---

## probability of Decision exchange

### process

---

### 示意圖
XXXXlast week testXXXX
VVVVthis weekVVVV

---

| time | neuron | input_str |
|:----------:| ------ | ---------- |
| 3000 | 0 | 8 |
| 3000 | 1 | 0 |
| next | | |
| ***var1*** | 0 | 0 |
| ***var1*** | 1 | ***var2*** |
| next | | |
| 3000 | 0 | 8 |
| 3000 | 1 | 0 |

## every var1 and var2 run 1000 times to cal those probability.

---

### time_400,700,1000

---

### detailtime400

---

## inverse spikes interval

---